Patient and operation related factors influencing the outcome of knee arthroplasty

(1)

DISSERTATIONS | JAAKKO JÄRVENPÄÄ | JULKAISUN NIMIKE, PITKÄ NIMIKE JATKUU NÄIN... | No 00

uef.fi

PUBLICATIONS OF

THE UNIVERSITY OF EASTERN FINLAND Dissertations in Health Sciences

ISBN 978-952-61-2312-7 ISSN 1798-5706

Dissertations in Health Sciences

THE UNIVERSITY OF EASTERN FINLAND

JAAKKO JÄRVENPÄÄ

PATIENT AND OPERATION RELATED FACTORS INFLUENCING THE OUTCOME OF KNEE ARTHROPLASTY

The continuous growth in the proportion of the population with obesity, combined with an increased demand for knee arthroplasty as the

population becomes increasingly elderly, will inevitably lead to a rise in the number of obese

patients requesting knee replacement surgery.

Knee arthroplasty surgery has progressed substantially in recent years but many technical

challenges are still inherent in this surgical procedure in obese or morbidly obese patients.

This entirety will pose a major challenge for orthopaedic surgeons in the coming years.

JAAKKO JÄRVENPÄÄ

(2)

(3)

JAAKKO JÄRVENPÄÄ

Patient and operation related factors

influencing the outcome of knee arthroplasty

To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in Auditorium 2, Kuopio University Hospital, Kuopio, on

Friday, November, 25^th 2016, at 12 noon.

Publications of the University of Eastern Finland Dissertations in Health Sciences

Number 383

Department of Orthopaedic, Traumatology and Hand Surgery of Kuopio University Hospital and Kuopio Musculoskeletal Research Unit, Institute of Clinical Medicine, School of Medicine, Faculty

of Health Sciences, University of Eastern Finland Kuopio

2016

(4)

Jyväskylä, 2016 Series Editors:

Professor Tomi Laitinen, M.D., Ph.D.

Institute of Clinical Medicine, Clinical Physiology and Nuclear Medicine Faculty of Health Sciences

Professor Hannele Turunen, Ph.D.

Department of Nursing Science Faculty of Health Sciences Professor Kai Kaarniranta, M.D., Ph.D.

Institute of Clinical Medicine, Ophthalmology Faculty of Health Sciences

Associate Professor (Tenure Track) Tarja Malm, Ph.D.

A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences

Lecturer Veli-Pekka Ranta, Ph.D. (pharmacy) School of Pharmacy

Faculty of Health Sciences Distributor:

University of Eastern Finland Kuopio Campus Library

P.O.Box 1627 FI-70211 Kuopio, Finland http://www.uef.fi/kirjasto ISBN (print): 978-952-61-2312-7

ISBN (pdf): 978-952-61-2313-4 ISSN (print): 1798-5706

ISSN (pdf): 1798-5714 ISSN-L: 1798-5706

(5)

Author’s address: Department of Orthopaedic, Traumatology and Hand Surgery Kuopio University Hospital

FINLAND

Supervisors: Docent Jukka Kettunen, M.D, Ph.D.

Department of Orthopaedic, Traumatology and Hand Surgery Kuopio University Hospital

KUOPIO FINLAND

Professor Heikki Kröger,M.D, Ph.D.

KUOPIO FINLAND

Docent Hannu Miettinen, M.D, Ph.D.

KUOPIO FINLAND

Reviewers: Professor Juhana Leppilahti, M.D, Ph.D.

Department of Orthopaedic and Trauma Surgery Oulu University Hospital

OULU FINLAND

Docent Jorma Pajamäki, M.D, Ph.D.

COXA Hospital University of Tampere TAMPERE

FINLAND

Opponent: Docent Petri Virolainen, M.D, Ph.D.

Department of Orthopaedics and Traumatology Turku University Hospital

TURKU FINLAND

(6)

(7)

Järvenpää, Jaakko

Patient and operation related factors influencing the outcome of knee arthroplasty University of Eastern Finland, Faculty of Health Sciences

Publications of the University of Eastern Finland. Dissertations in Health Sciences 383. 2016. 65 p.

ISBN (print): 978-952-61-2312-7 ISBN (pdf): 978-952-61-2313-4 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706

ABSTRACT:

It has been shown that obesity increases general risk factors for complications and poorer outcome scores after total knee arthroplasty (TKA) although conflicting findings have been reported concerning the relationship between obesity and TKA. There is a risk for the patient to need revision surgery after unicompartmental knee arthroplasty (UKA) and conversion to TKA is often necessary. Periprosthetic femoral bone mineral density (BMD, g/cm2) has been shown to decline after TKA but the effect of this stress-shielding related bone loss on clinical outcome is not properly understood.

The main aims of this doctoral thesis were to examine the effect of obesity both on (I) short and (II) long term results after the TKA operation including postoperative complications, knee and functional outcome scores and patient-assessed satisfaction. Long- term results of failed UKAs which had been converted to TKA were evaluated and their results were compared to the results of a primary TKA operation (III). The last study (IV) examined the long-term changes of the bone mineral density of the distal femur after TKA.

We found that obesity exerted a negative impact on the short and long-term outcome after TKA as assessed by patients’ clinical function and satisfaction (ROM, KSS, and WOMAC). UKA conversion to TKA was associated with poorer clinical outcome as compared to primary TKA as reflected in both patients’ functional level and satisfaction (ROM, WOMAC). The mean of periprosthetic bone loss during the 7-year follow-up varied from 10.3% to 30.6% depending on the region of interest. The highest bone loss rates were detected during the first three postoperative months but periprosthetic bone loss around the femoral component continued for up to 7 years after the TKA operation. However, the loss of bone density was not associated with any negative clinical outcome or complications during the follow-up.

National Library of Medicine Classification: WE 874, WD 210

Medical Subject Headings: Arthroplasty, Replacement, Knee; Arthritis; Bone Density; Follow-Up Studies;

Knee Joint; Obesity; Personal Satisfaction; Postoperative Complications; Postoperative Period; Rehabilitation;

Treatment Outcome

(8)

Järvenpää, Jaakko

Polven tekonivelkirurgian tuloksiin vaikuttavat potilas – ja operaatiokohtaiset tekijät Itä-Suomen yliopisto, terveystieteiden tiedekunta

Publications of the University of Eastern Finland. Dissertations in Health Sciences 383. 2016. 65 s.

ISBN (print): 978-952-61-2312-7 ISBN (pdf): 978-952-61-2313-4 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706

TIIVISTELMÄ:

Lihavuuden on osoitettu lisäävän polven tekonivelleikkauksiin liittyviä riskejä ja vaikuttavan myös leikkauksen jälkeiseen kuntoutumiseen. Tulokset ovat edelleen silti ristiriitaisia. Polven osatekonivelleikkaukseen liittyy suurentunut riski uusintaleikkaukselle, jolloin tilalle vaihdetaan useimmiten kokotekonivel. Luun mineraalitiheys polven tekoniveltä ympäröivässä luussa heikkenee tekonivelleikkauksen jälkeen. Sen merkitys polven tekonivelleikkauksen tuloksiin on kuitenkin edelleen epäselvää.

Tämän tutkimuksen tarkoituksena oli selvittää (I) lyhyellä ja (II) pitkällä aikavälillä lihavuuden vaikutuksia polven tekonivelleikkauksen jälkeiseen kuntoutumiseen arvioimalla potilaiden toimintakykyä ja tyytyväisyyttä. Tavoitteena oli myös arvioida samoilla menetelmillä, pitkän seuranta-ajan kuluessa, polven osatekonivelen uusintaleikkauksiin liittyviä mahdollisia haittoja (III). Viimeisessä osatyössä (IV) selvitettiin distaalisen reisiluun periproteettisessa luussa tapahtuvaa mineraalitiheyden heikkenemistä pitkän seurantavälin aikana.

Tulostemme mukaan lihavuus saattaa huonontaa leikkauksen jälkeistä kuntoutumista ja potilaiden tyytyväisyyttä. Polven liikettä, kipua ja potilaan toimintakykyä mittaavilla menetelmillä (ROM, KSS, WOMAC) lihavien potilaiden tulokset olivat heikommat. Polven osatekonivelen uusintaleikkauksiin liittyi myös riski polvinivelen huonommasta liikelaajuudesta ja potilaan toimintakyvystä leikkauksen jälkeen. Periproteettinen reisiluun alaosan mineraalitiheys vähenee suurimmaksi osaksi heti ensimmäisten kolmen leikkauksen jälkeisen kuukauden aikana. Luun mineraalitiheyden heikkeneminen jatkuu tämän jälkeen edelleen jopa seitsemän vuoden kohdalla leikkauksen jälkeen, mutta selvästi hitaammin. Periproteettisen luun heikkenemiseen ei liittynyt tässä tutkimuksessa potilaan toimintakykyyn liittyviä ongelmia tai komplikaatioita.

Yleinen Suomalainen asiasanasto: kuntoutuminen, leikkaushoito, lihavuus, luuntiheys, nivelrikko, polvet, potilaat, proteesit, riskitekijät, tekonivelet

(9)

To Sanna, Viljami and Helmi

(10)

Acknowledgements

This research was carried out in the Department of Orthopaedic, Traumatology and Hand Surgery of Kuopio University Hospital and Kuopio Musculoskeletal Research Unit in University of Eastern Finland during years 2009-2016. I owe my deepest gratitude to all the people who have contributed to this work.

Especially I would like to express my deepest respect and thanks to:

Docent Jukka Kettunen, my principle supervisor for his warm support and excellent guidance in this project as well in many other aspects in the field of orthopaedics. You originally encouraged me to start this work and supported me to persist with it during these hectic years.

Professor Heikki Kröger, my second supervisor, for providing me with the opportunity to join this research team. I also admire your exceptional and endless scientific expertise and knowledge which helped me to complete this study process.

Docent Hannu Miettinen, the Chief of Department of Orthopaedic, Traumatology and Hand Surgery, my third supervisor, for his valuable comments, feedback and support.

Tarja Soininvaara, M.D., my technical supervisor, for her skillful guidance in DXA- methodology, passionate attitude towards scientific research and especially her co- operation in writing the manuscripts.

Docent Jorma Pajamäki and Professor Juhana Leppilahti, my official reviewers, for their valuable comments and corrections.

Nurse researchers Mrs Elina Jalava and Riitta Toroi for their practical help in polyclinic patient work. Ewen MacDonald, Ph.D. for revising the English language of this thesis.

Maija Pesola, M.D., the Chief of Department of Orhopaedic of Central Finland Central Hospital and Orthopedists Kati Kyrölä and Konsta Pamilo from Central Finland Central Hospital and Orthopedist Jussi Jalkanen from Kuopio University Hospital for their support and advices since my first steps of my career in surgery and orthopaedics.

All my friends and colleagues, especially Antti, Atte, Henrik, Tommi, Tuomas and Valtteri for their continuous support since the first day of my career in medicine. Jaakko, my childhood friend, thank you for being my ‚brother‛, friend and confidant during these 20- years.

My parents Pirkko and Jouko for their encouragement and support throughout my life.

And finally, I owe my deepest and loving thanks to my wife Sannamaaria and my lovely children Viljami and Helmi for being my greatest treasures.

(11)

For financial support, I would like to thank The Finnish Medical Society Duodecim, The Finnish Arthroplasty Society and Kuopio University Hospital.

Kuopio, September 2016 Jaakko Järvenpää

(12)

List of the original publications

This dissertation is based on the following original publications, which are referred to in the text by their Roman numerals:

I Järvenpää J, Kettunen J, Kröger H, Miettinen H. Obesity may impair the outcome of total knee arthroplasty. A prospective study of 100 patients. Scand J Surg 99:45−49, 2010.

II Järvenpää J, Kettunen J, Soininvaara T, Miettinen H, Kröger H. Obesity has a negative impact on clinical outcome after total knee arthroplasty. Scand J Surg 101: 198−203, 2012.

III Järvenpää J, Kettunen J, Miettinen H, Kröger H. The clinical outcome of revision knee replacement after unicompartmental knee arthroplasty versus primary total knee arthroplasty: 8-17 years’ follow-up of 49 patients. Int Orthop 34: 649−653, 2010.

IV Järvenpää J, Soininvaara T, Kettunen J, Miettinen H, Kröger H. Changes in bone mineral density of the distal femur after total knee arthroplasty: A 7-year follow- up comparing results between obese and nonobese patients. Knee 21: 232-235, 2014.

The publications were adapted with the permission of the copyright owners.

(13)

(14)

Abbreviations

AP Antero-posterior

ASO Atherosclerosis obliterans BMD Bone mineral density BMI Body mass index CR Cruciate retaining

DXA Dual Energy X-ray absorptiometry ENMG Electroneuromyography

FS Function score

KOOS Knee injury and osteoarthritis outcome score

KS Knee score

KSS Knee society score

MCL Medial collateral ligament OA Osteoarthritis

OKS Oxford knee score PS Posterior stabilized ROM Range of motion ROI Region of interest SD Standard deviation TKA Total knee arthroplasty

UKA Unicompartmental knee arthroplasty US United States

VAS Visual analog scale

WHO World Health Organization

WOMAC Western Ontario and McMaster Universities osteoarthritis index

(16)

1 Introduction

The knee joint is the largest weight bearing joint in the human body and knee osteoarthritis (OA) is their commonest joint disorder all around the world. Today, knee joint replacement surgery has become a reliable and standard procedure for the treatment of advanced knee OA.

Overweight and obesity are becoming increasingly common. The global rate of obesity has more than doubled since 1980 (WHO 2015). Obesity has been linked to the development of knee OA and it increases the probability of TKA. On the other hand, obesity is also associated with a poorer outcome after TKA. Conflicting findings have been reported concerning the relationship between obesity and TKA (Felson et al. 1988, Foran et al. 2004, Gunst and Fessy 2015, Harms et al. 2007, Spicer et al. 2001). Although body mass index (BMI) as an indicator of obesity is thought to be a negative prognostic factor in TKA operations, the relationship between the BMI and postoperative outcome following TKA is still controversial (Deshmukh et al. 2002, Lozano et al. 2015, Rodriquez-Merchan 2015, Spicer et al. 2001).

The annual number of the knee arthroplasty operations has been increasing rapidly during the last 20 years and the demand for this procedure is expected to grow in the future. It also seems likely that the increased demand for knee replacement surgery will also concentrate to obese patients. It must be noted that since there will be increases in the number of total knee replacement patients, the number of major complications requiring a revision operation also will grow in the future (Cornell 2015).

Due to the continuous development of surgical techniques and improvements in component designs since the early 1970s, unicompartmental knee arthroplasty (UKA) has become a more successful and reliable treatment method for unicompartmental knee osteoarthritis. Several studies have claimed that UKA is a feasible option for the treatment of unicompartmental osteoarthritis. Even excellent outcomes have been reported when focusing attention on careful patient selection and correct indications (Berger et al. 2005, Griffin et al. 2007, Insall and Walker. 1976). The advantages of UKA in comparison to TKA are its cost-effectiveness, fewer serious complications and better postoperative functional outcome (Berger et al. 2005, Lim et al. 2012, Niinimäki et al. 2014, Robertsson et al. 1999).

However, an increased risk of revision has been shown to be a disadvantage of this method. When there is a failure of UKA, a revision procedure to TKA is often necessary.

Some authors have reported the outcomes of UKA revisions to be comparable to those achieved with primary TKA but better than those of revised TKA (Ackroyd et al. 2003, Levine et al. 1996, Lunebourg et al. 2015). Despite concerns regarding a higher risk of revision, UKA still continues to be utilized as an alternative to total knee arthroplasty.

When choosing between the UKA and TKA, patients should be informed of the advantages of both procedures, but they also should be advised about the generally higher revision risk after UKA (Lim et al. 2012, Niinimäki et al. 2014).

(17)

Dual-energy X-ray absorptiometry (DXA) is a reliable method for measuring bone mineral density (BMD, g/cm²). BMD is an important measure of bone strength and quality.

The application of DXA has been gradually extended from the initial studies of osteoporosis to different areas of interest in orthopaedic surgery. Nowadays, DXA measurement provides a highly reproducible method for quantitative analysis and monitoring periprosthetic BMD after TKA and adaptive bone remodeling of both the proximal tibia and distal femur occurs. Several studies have shown that this process will result in a marked reduction in BMD. This decline in BMD seems to occur most rapidly during the first 3 postoperative months. However long-term periprosthetic bone loss have been less extensively studied or published (Cavalli and Brandi 2014, Parchi et al. 2014, Robertson et al. 1994, Soininvaara 2005).

All these patient- and operation-related factors influencing the outcome of knee arthroplasty are very complex. In this doctoral thesis, we tried to assess the effect of obesity, operation technique and periprosthetic BMD on the outcome of knee replacement surgery.

(18)

2 Review of the literature

2.1 KNEE JOINT AND KNEE OSTEOARTHROSIS

2.1.1 Development and risk factors of the osteoarthritis (OA)

Knee osteoarthritis is the most common joint disorder causing pain and disability especially in the elderly population. OA is a dynamic and metabolically active process affecting the whole joint involving the destruction of cartilage, the synovium and also extending to the extra-articular muscles (Arokoski, et al. 2007, Brandt et al. 2006).

Knee OA is a multifactorial disease and the actual initiating cause of the pathological process remains unknown. However our present understanding suggests that not only atypical mechanical loading but also normal loading of structurally weakened cartilage tissue may be responsible for a biomechanical process leading to the destruction of cartilage and thus further to OA (Arokoski et al. 2000, Moskowitz et al. 2007, Van der Kraan and Van der Berg 2012).

The most important systemic and biomechanical risk factors leading to OA are aging, obesity, knee joint trauma, joint malalignment and excessive physical activity. Aging is the most important individual risk factor for OA but obesity significantly increases the need for a possible arthroplastic procedure (Arokoski et al. 2007, Derman et al. 2014, Doherty 2001, Felson et al. 1988, Sharma et al. 2000, Sharma et al. 2001).

Normal anterior-posterior and sagittal X-ray pictures have traditionally been used to evaluate the grade of preoperative OA of the knee joint and postoperative radiological outcome after knee arthroplasty. Kellgren and Lawrence (1957) were the first investigators to provide a classification where OA was divided into the five grades (None, Doubtful, Minimal, Moderate, Severe) depending on the severity of joint destruction.

Varus and valgus alignment are associated with the progression of knee osteoarthritis, but their overall role in OA disease is not totally clear. It has been shown that nearly one in every three patients may have OA in the medial compartment only and an isolated lateral compartment OA of the knee is less common (Ledingham et al. 1993, Sharma et al. 2001, Sharma et al. 2013).

(19)

2.2 KNEE ARTHROPLASTY

2.2.1 Total knee arthroplasty (TKA)

OA of the knee can severely impact on a patient’s functional status and quality of life.

Total knee arthroplasty (TKA) is a reliable and one of the most cost-effective methods for the treatment of knee osteoarthritis. TKA is primarily indicated in severe degenerative arthritis and inflammatory arthritis as well in post-traumatic arthritis when conservative methods are no longer effective. The aims of this well-characterized surgical procedure are not only to achieve pain relief but also to improve knee function and the total quality of life (Arokoski et al. 2007, Ethgen et al. 2004, Gill and Joshi. 2001, Koskinen et al. 2008).

In the primary TKA operation, the damaged articular cartilage and subchondral bone are removed and replaced by metallic (chromium-cobalt and titanium) and plastic (polyethylene) prosthetic components which are designed to mimic the normal anatomical joint shape and function. Successful total knee arthroplasty has often been assessed on whether it achieves the restoration of the knee to neutral alignment postoperatively. Due to the continuous development of surgical techniques and prosthesis designs occurring over last decades, the TKA procedures have become increasingly reliable treatment protocols (Cherian et al. 2014, Koskinen et al. 2008, Luo 2004).

The annual number of the knee arthroplasty surgical operations has been increased dramatically during the last 20 years and the demand is expected to grow also in the future. The most substantial change has occurred in the numbers of elderly patients but in addition, there has been a substantial growth in the numbers of TKA operations performed on patients aged 50 − 60 years (Derman et al. 2014, Dixon et al. 2004, Robertsson al. 2000, Robertsson et al. 2001).

Since the 1980s, the annual volume of TKAs has increased significantly in Finland (Leskinen et al. 2012), for example in 2013, a total of 11433 knee arthroplasty operation were performed in the nation’s hospitals (National Institute of Health and Welfare 2013).

The survival of the knee prostheses has improved during the most recent decades; in several register studies the 5 to 15 years’ survival rates of prostheses have been reported to be between 88 and 98 % when using the need for some kind of revision as an endpoint ( Argenson et al. 2013, Koskinen et. al 2008, Niinimäki et al. 2014, Victor et al. 2014).

2.2.2 Unicompartmental knee arthroplasty (UKA)

Since the early 1970s, UKA has also become a more popular method for treating unicompartmental medial condylar knee osteoarthritis. However, there is still no consensus about its benefits as a surgical procedure (Insall and Walker. 1976, Koskinen et al. 2007, Koskinen et al. 2008).

At the present time, UKAs are widely used to treat isolated unicompartmental knee osteoarthritis although it is important to ensure careful patient selection and correct surgery-related indications such as stable knee joint, no high-grade deformity and one

(20)

compartment disease. It seems that the advantages of UKA include lower postoperative morbidity, a quicker return to routine activities, and more normal feelings in the knee (Berger et al. 2005, Lim et al. 2012, Niinimäki et al. 2014). In their systematic review, Griffin et al. (2005) found that the range of motion was better in UKA compared with TKA.

Complication rates after UKA and TKA were similar but deep vein thrombosis was reported more often after TKA. The functional outcome after UKA appears to be at least as effective as TKA.

It is believed that UKA has better bone sparing properties and relies more on an intact ligament-based stability compared to TKA. However, there are similarities between the procedures i.e. both implants replace the articular surfaces of either the medial or the lateral femoral condyle and the adjacent tibial plateau surface (Canale and Terry 2013).

It has been reported that there is a greater risk of revision following UKA compared with primary TKA (Niinimäki et al. 2011). This is estimated via a term called survivorship;

this is the most important end point after knee replacement surgery. According to a large register study based data from Finnish Arthroplasty Register, it seems that UKAs had an inferior long-term survivorship compared with cemented TKAs, even after adjusting for the age and sex of the patients. Kaplan-Meier survivorship of UKAs was 89.4% at 5 years, 80.6% at 10 years, and 69.6% at 15 years; the corresponding rates for TKAs were 96.3%, 93.3%, and 88.7% (Niinimäki et al. 2014).

UKA and TKA are both recommended for the treatment of medial compartment osteoarthritis in the varus knee. Traditionally, studies have reported favorable functional results and patient satisfaction from TKA i.e. this knee arthroplastic procedure seems to be offer long-term success (McAllister 2008). There are also studies reporting that UKA may confer promising advantages over TKA if one takes into consideration cost-effectiveness, speed of recovery, postoperative range of motion and the retention of the cruciate ligaments (Arirachakaran et al. 2015, Padgett et al. 1991, Soohoo et al. 2006,).

2.2.3 UKA revision

Some surgeons have reported that outcomes after UKA revision can be comparable to TKA (Chakrabarty et al. 1998, Foong and Lo 2014). Nevertheless, it has also been generally observed in long term follow-up studies that the revision UKA to TKA is technically more difficult and functionally less satisfactory in terms of the final follow-up when compared to situation after a primary TKA (Craik et al. 2015, Rancourt et al. 2012).

When a UKA failure occurs, a revision procedure will usually be needed i.e. UKA needs to be converted to TKA. Despite concerns regarding a higher risk of revision, UKA continues still to be used as an alternative to TKA. When choosing between the UKA and TKA, patients should be informed of advantages of both procedures; in particular, they also should be advised about the generally higher revision risk after UKA (Koskinen et al.

2008, Niinimäki et al. 2011, Niinimäki et al. 2014).

(21)

2.2.4 Measuring outcome after knee replacement surgery

As is the case with many medical procedures, the evaluation of parameters like value and effectiveness of TKA depend on our definition of a ‘successful’ treatment. These include relief of the patient’s symptoms, restoration of their physical function, and improvement of their post-operative condition from the pre-operative situation. Previous studies have shown that somewhere from 10% to 25% of patients are dissatisfied with the outcome of knee replacement at one to three years after surgery. The main reason is the presence of residual symptoms and the failure of the procedure to meet the patient’s preoperative expectations (Gandhi et al. 2009, Lingard et al. 2006, Noble et al. 2005, Noble et al. 2006). The presence of preoperative psychologic distress has been also associated with poorer 1-year outcomes for function and quality of life in patients undergoing TKA (Utrillas-Compaired et al. 2014).

Although the patients’ subjective opinions are the most valuable after operation, other parameter measuring protocols including both clinical and radiological definitions of outcome have been used to estimate the results after knee replacement surgery. In general, a deviation from the normal mechanical axis of less than ± 3 degrees from neutral has been considered to be an acceptable outcome measured from postoperative anterior-posterior radiographs (Fig. 2) (Ewald 1989, Ramkumar et al. 2015, Sampath et al. 2009).

Patients’ reported outcome measures are widely used in the assessment of outcomes after TKA in joint registries and large studies. The approach of the different measurements protocols can vary but all need to be valid, reliable and responsive (Kreibich et al. 1996).

There are four commonly used objective and subjective protocols for evaluating the quality of life and ability to function after knee replacement surgery: Knee Society Score rating system (KSS) (Insall et al. 1989), Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) (Bellamy et al. 1988, Ramkumar et al. 2015), Oxford Knee Score (OKS) (Maempel et al. 2016) and Knee Injury and Osteoarthritis Outcome Score (KOOS) (Collins et al. 2016).

The WOMAC include 24 questions in three classes (Pain, Stiffness and Physical function). The patient is given a VAS (visual analogue scale)-assessment of the WOMAC on a scale from 0 mm (no pain, stiffness or disability in physical function) to 100 mm (severe pain, stiffness or disability on physical function), and the sum of scores is calculated within all classes. The aggregate scores for each dimension are calculated as an average within all three classes.

The postoperative walking ability and balance can be measured in a timed Get- Up- and- Go test. In this test, the patient is initially sitting on a chair and stands up after receiving a sign, walks 3 meters and then turns around, returns to the chair and sits down.

The time from leaving the chair until returning back and being seated again is measured.

This test assesses the patient’s postoperative walking ability and balance (Mathias et al.

1986).

Patients’ preoperative and postoperative pain can be assessed reliably by levels from 1 to 10 on a Visual Analog Scale (VAS) (Hawker et al. 2011).

(22)

Matsuda et al. (2013) concluded in their retrospective questionnaire that patient satisfaction is difficult to measure. The avoidance of varus alignment and the achievement of better ROM appear to be important for increasing patient satisfaction.

Figure 1. The mechanical axis of the lower limb is defined by combining the mechanical axis of the femur defined by the line from the centre of the femoral head to the centre of the knee joint and the mechanical axis of the tibia defined by the line from the centre of the knee to the centre of the ankle. A deviation of within ± 3 degrees from neutral is considered to be an acceptable outcome.

(23)

2.3 KNEE ARTHROPLASTY AND OBESITY 2.3.1 Definition and general risk factors of obesity

Overweight and obesity are defined as abnormal or excessive fat accumulation that presents a risk to health (Bray 1987, Lahti-Koski et al. 2010). A crude population measure of obesity is the body mass index (BMI) in which a person’s weight (in kilograms) is divided by the square of his/her height (in metres). The World Health Organization (WHO) classifies obesity also on the basis of BMI. A value of BMI of 30 (kg/m²) or more is generally considered as obese. A person with BMI ≥ 25 is considered to be overweight.

Those persons with BMI ≥ 40 are morbidly obese (WHO 2004).

Obesity has reached epidemic proportions all around the world. According to WHO (2015), worldwide obesity has more than doubled since 1980 and most of the world's population live in countries where overweight and obesity kill more people than underweight and malnutrition. There are epidemiological studies emphasizing that overweight and obesity are major risk factors not only for OA but also for a number of chronic diseases including cardiovascular disease, type 2 diabetes, hypertension, cancer and premature death. The high prevalence of obesity, combined with its concomitant cardio-metabolic risks, means that it is now a major global health challenge (Abelson and Kennedy 2004, Haslam and James. 2005, Nathan 2015). In recent dynamic mathematic model study, it has been estimated that the prevalence of obesity in the United States (US) appears to be levelling off, but the reasons behind this plateau situation remain unknown.

This model predicts that by the year 2030 there will be the following prevalence rates;

overweight - 28%; obesity - 32% and extreme obesity - 9% (Thomas et al. 2014).

2.3.2 Limitations of using BMI to measure obesity

The relationship between BMI and body fatness varies according to body composition proportions (Garn et al. 1996). For example, the percentage of body fat mass is higher in women than in men with similar BMI values. BMI is also known to vary with age, therefore the age group under examination will affect the results on the prevalence of obesity. In practice, body fatness has been noted to increase with aging, meaning that BMI may correspond to a greater body fat content in elderly persons compared with younger individuals (Ross et al. 1994, Gallagher et al. 1996).

The age-related change in height can also influence BMI. In adults, height is lost with normal aging. There is an average height loss of 3 cm occurring from age 30 to 70 years and this has been estimated to account for an artificial increase in BMI of 0.7 kg/m2 for men. In women, the height loss averaged about 5 cm over the same 40-year period, accounting for an increase of 1.6 kg/m2 in BMI (Sorkin et al. 1999).

It has also been shown that young athletes who may have higher body weight due to higher lean body mass, may seem to be misclassified as obese according to their BMI value (Jonnalagadda et al. 2004). On the other hand, patients undergoing knee arthroplasty are often elderly (Koskinen et al. 2008) and presumably not very athletic (Dahm et al. 2008).

(24)

Thus, despite its limitations, BMI provides a simple, useful and widely utilized population level measure of obesity in adults (WHO 2000).

The general WHO classification of BMI is shown in table 1.

Table1. WHO classification of body mass index. Modified from WHO Global database on body mass index 2015.

WHO Classification of BMI WHO BMI cut-off points kg/m²

Under weight < 18.5

Normal range 18.5 – 24.9

Overweight ≥ 25.0

Pre-obese 25.0 – 29.9

Obese class I 30 – 34.9

Obese class II 35.0 -39.9

Obese class III ≥ 40

2.3.3 Obese patients with TKA

The prevalence of obesity has increased during the past two decades in Europe and the adverse trends in obesity indicators have continued also in Finland in the 21^st century.

(Berghöfer et al. 2008, Lahti- Koski et al. 2010, Lahti-Koski et al. 2012). A recent study from the United States reported age-adjusted obesity prevalence to be 32% for men and 36% for women (Odum et al. 2013). Obesity is a major risk factor for osteoarthritis of the knee (Felson et al. 1988). Furthermore, osteoarthritis limits activity leading easily to weight gain and therefore obese patients will also be over-represented in the patient population who are candidates for knee arthroplasty surgery (Böstman 1994). In two studies originating from United States, over half of the total knee arthroplasty participants were obese (Namba et al. 2005, Fehring et al. 2007).

2.3.4 Obesity and TKA outcome

Obesity has been linked to the development of osteoarthritis of the knee and since the incidence of obesity is increasing, the need for total knee arthroplasty (TKA) is likely to increase (Fehring et al. 2007, Harms et al. 2007). Higher BMI also leads to joint replacement treatment in younger age (Guenther et al. 2015).

(25)

The increased demand for knee replacement surgery for obese patients would be predicted to lead to an increase in the number of patients experiencing major complications some of which may even require a revision operation (Cornell 2015).

Although the BMI is widely considered to be a negative factor in TKA operations, the relationship between the BMI and postoperative outcome following TKA is still controversial. Despite the increasing probability of wound infections, risk of ligament injury and technical difficulties encountered during the operation, satisfactory results have been reported also for obese patients (Spicer et al. 2001, Lozano et al. 2015). Deshmukh et al. (2002) studied a group of 180 TKA patients; in their short 1 –year follow-up analysis it was stated that BMI as a measure of obesity had no adverse influence on the outcome of TKA. However, Rodriquez-Merchan (2015) conducted a meta-review of 41 articles and he concluded that 16 studies reported no adverse association between obesity and TKA outcome but in 24 studies, it seemed that obese patients experienced a poorer TKA outcome.

There is extensive evidence that patients with a BMI ≥ 30 kg/m² are at a higher risk of exhibiting a lower range of motion, poorer functional scores, more dissatisfaction and a higher risk of developing complications following primary TKA. For this reason, obese patients are encouraged to lose weight before the TKA procedure (Shoji et al. 1990, Namba et al.2005, Amin et al. 2006).

TKA in morbidly obese patients has been associated with an increased rate of perioperative complications, including problems with wound healing, infection, and avulsion of the MCL. It has also been demonstrated that revision rates after primary total knee arthroplasty tend to be higher in obese patients (Winiarsky et al. 1998, Foran et al.

2004, Kerkhoffs et al. 2012, Gunst and Fessy. 2015). Jämsen et al. (2010) evaluated their large and unselected consecutive 2,647 knee arthroplasty series and found that patient- related factors such as comorbidity, poor preoperative clinical state and obesity were risk factors for postoperative prosthetic infections during the first year after the operation when total rates of superficial and deep prosthetic joint infections were 2.9 % and 0.8 % respectively.

Lübekke et al. (2016) compared non-obese and obese patients in their long-term follow- up study. This follow-up investigated a total of 3,438 TKA patients. In that study, BMI ≥ 35 (Obese class II) and body weight more than 100 kg were identified as the threshold for a significant increase in the early postoperative infection rate. They also suggested that BMI ≥ 35 or weight ≥ 100 kg may serve as a cutoff for higher perioperative dosage of antibiotics.

Super-obesity (BMI > 50 kg/m²) has also been described as a BMI subgroup in the literature. Werner et al. (2015) noticed that super-obesity was associated with dramatically increased rates of postoperative local and systemic complications after TKA compared to non-obese, obese and even morbidly obese patients.

(26)

2.4 BONE MINERAL DENSITY AND KNEE ARTHROPLASTY 2.4.1 Dual-energy x-ray absorptiometry (DXA)

Bone mineral density (BMD) reflects the amount of bone mineral in bone tissue (g/cm2).

Dual-energy x-ray absorptiometry (DXA) is a reliable means to measure BMD. In DXA measurements, two X-ray beams with different energy levels penetrate through the bone tissue. When soft tissue absorption is subtracted from the total, the BMD can be determined from the bone absorption of each beam (Cullum et al. 1989, Soininvaara 2005).

Bone mineral density is an important measure of bone strength and quality. The application of DXA in orthopaedic surgery has gradually been extended from the initial studies conducted in patients with osteoporosis to different areas of interest. Nowadays, the DXA technique provides a highly reproducible method for quantitative analysis and monitoring periprosthetic BMD after total knee arthroplasty. DXA has also been shown to be a many times more accurate method of evaluating bone quality than visual inspection from X-ray pictures (Lewis et al. 1998, Cavalli and Brandi 2014).

The wide range of prosthesis designs and models and the increasing demand of knee replacement surgery will also increase the various types of complications. Therefore, knowledge of the potential complications and their imaging appearance will be helpful in the diagnostic evaluation of a patient with a painful knee arthroplasty (Miller 2005).

2.4.2 Bone mineral density after knee arthroplasty

Postoperative X-rays can be used to assess the implant position, the knee alignment, to evaluate the bone-implant interface and reveal the presence of knee infection or implant loosening. In most cases, plain radiographs are of little benefit when evaluating the quantitative changes in periprosthetic bone especially those occurring during the early stages after the operation. The poor sensitivity of plain radiography has led to its replacement by DXA as the best way to quantify periprosthetic bone loss after knee arthroplasty and at the present time, DXA represents a highly reproducible method for quantitative analysis and monitoring periprosthetic BMD after TKA (Robertson et al. 1994, Liu et al. 1995, Soininvaara et al. 2000, Soininvaara 2005, Parchi et al. 2014).

The introduction of a knee implant inevitably alters the physiological transmission of load to the surrounding bone and initiates an adaptive bone remodeling process, resulting in a reduction in the periprosthetic bone mineral density. After TKA, there is an adaptive bone remodeling of both the proximal tibia and distal femur. Several studies have shown that most often this process seems to result in a reduction in BMD. Bone remodeling is defined as regional removal of old bone and its replacement with newly formed bone.

Stress shielding as a consequence of altered biomechanical loading at the periprosthetic areas and also bone reaction to surgical trauma, and the brief postoperative period of immobilization of the extremity are considered to be the major causes of early phase bone loss. In theory, this could lead to migration, instability and aseptic loosening of the prosthetic components (Cavalli and Brandi 2014, Karbowski et al. 1999, Parchi et al. 2014, Petersen et al. 1996, Soininvaara et al. 2005).

(27)

Changes in bone mineral density have been a topic of academic interest for the last two decades. It has been shown in previous short-term studies that the periprosthetic BMD decrease will continue for at least 12 months after the TKA operation. This decline in bone density seems to be most rapid during the first 3 months, and subsequently the loss in BMD still continues but at a slower rate. It has also been reported that the average bone density loss has ranged from 11 to 15.6% in all metaphyseal areas during first three postoperative months. After 12 months, the mean bone-density loss has been estimated to lie in a range from 12.1 to 22.8%. At the tibial level, the baseline BMD of the medial region of interest (ROI) was higher in the varus knees than in the valgus aligned knees Liu et al.

1995, Petersen et al. 1996, Soininvaara et al. 2004).

(28)

3 Aims of the study

The aims of the present study were to:

1. Examine the effect of obesity on short-term results after a TKA operation including an evaluation of postoperative complications and functional outcome (I).

2. Determine the effect of obesity on long-term results after TKA including an evaluation of postoperative complications and functional outcome (II).

3. Evaluate the long-term results of a failed UKA which needs to be converted to TKA in comparison to the results achieved in a primary TKA operation (III).

4. Assess the long-term changes of BMD in the distal femur between obese and non- obese patients after TKA (IV).

(29)

4 Materials and methods

4.1 GENERAL STUDY DESIGN

The total population examined in the present thesis consists of 234 patients from four subgroups who had undergone TKA or UKA in Kuopio University Hospital during the time period from 1991–2007.

The patients’ characteristics including age, gender, medication, smoking, preoperative body mass index (BMI) and postoperative hospital stay were recorded from the medical case records. All the operations were carried out in the Department of Orthopaedics and Traumatology in Kuopio University Hospital and with the standard surgical technique and all prostheses were implanted in bone cement (Palacos cum gentamycin, Schering- Plough, US). All patients received 3g cefuroxime at induction, 40 mg/kg tranexamic acid intravenously immediately after the operation and 40 mg sodium enoxaparin ten days’

postoperatively.

The duration of the operation, post-operative bleeding and transfusions were recorded.

Chronic diseases of the patients (ASO, hypertension, diabetes, cancer, or cardiac illness), medication, smoking, preoperative range of motion and the need for ambulatory support for walking (e.g. crutches, walker) were recorded. Long, weight-bearing coronal and sagittal radiographs of the operated limb were examined. The alignment of the knee was evaluated, and the degree of the deviation of the mechanical axis was defined as the angle between a line from the femoral head to the centre of the knee joint and a line from the centre of the knee to the centre of the ankle. A deviation of less than ± 3 degrees from neutral was considered to be an acceptable outcome. The operations were performed using a tourniquet and spinal anesthesia.

The approval of the local ethics committee was sought and obtained separately for each single study project.

4.1.1 BMI and short-term outcome after TKA (Study I)

Between October 2006 and March 2007, 100 primary unilateral TKA were performed in Department of Orthopedics and Traumatology, Kuopio University Hospital. The NexGen, posterior stabilized (PS) total knee prosthesis system (Zimmer, Warsaw, IN, USA) was implanted in 53 patients. The Triathlon cruciate retaining (CR) (Stryker Howmedica Osteonics, Mahway, NJ, USA) system was used in 46 patients and one operation was performed with Thriathlon posterior stabilized components. Each patient had a drain installed postoperatively. A total of 14 patellar surfacing (6 in Group 1 vs. 8 in Group 2) and 13 lateral releases (9 in Group 1 vs. 4 in Group 2) were performed. The consecutive patients were examined by an orthopedist approximately two weeks before the operation when their suitability for the procedure was evaluated.

(30)

The patients were invited to the Kuopio University Hospital for a control evaluation on an average at 3 months postoperatively. The patients were examined with a routine clinical procedure. The length of the wound was measured with a measuring tape with the knee as straight as possible, and ROM was measured with goniometry. Any postoperative complications were recorded. The level of pain was documented by asking the patients to rate it with a VAS. The patients' satisfaction with the operation was asked with a question, whether they would be willing to undergo again the same operation in the same situation: possible answers were ‚yes‛ or ‚no‛. The postoperative walking ability and balance were measured with a timed Get- Up- and- Go test. The mechanical axis of the lower limb was measured from the long, weight-bearing coronal and sagittal radiographs of the operated limb. A deviation of less than ± 3 degrees from neutral was considered to be a good outcome (Sampath et al. 2009).

The patients were divided into two groups (obese and non-obese) according to body mass index. Group 1 (non-obese, n = 48) consisted of patients whose body mass index was

< 30 kg/m ² and the patients with BMI ≥ 30 kg/m² made up Group 2 (obese, n = 52). The mean body mass index in Group 1 was 26.1 kg/m² (22.2 − 29.4); in Group 2 it was 33.0 kg/m² (30.0 − 40.5). The patients’ demographic data are shown in table 2.

The mean of the Kellgren-Lawrence classification was 3.1 (3.0−3.3) for the entire study population. There was no difference between the study groups in the radiological degree of severity of the osteoarthritis or the mechanical axis of the lower limb before operation:

KL: 3.2/3.1, p = 0.30; Mech. axis 8.0°/6.9°, p = 0.83. The preoperative deviation from the straight mechanical axis was at least 3 degrees in 39 patients in Group 1 and in 45 patients in Group 2, p = 0.46.

Table 2. Demographic data of the patients in study I expressed as number and mean (range).

Group 1(n = 48) (BMI < 30)

Group 2 (n = 52) (BMI ≥ 30)

p-value Men/women (n)

Age (years)

Chronic disease (yes) (n)^a Medication (yes) (n)^b Smokers (n)

Use of ambulatory support (preop.) (n)

20/28 68.8 ( 53−86) 34

4 8 17

14/38 67.4 ( 40−79) 35

5 7 25

0.12 0.42 0.7 0.82 0.65 0.2

a ASO, hypertension, diabetes, cancer and cardiac illness.

b Glucocorticoids, salazopyrin and cytostatics.

(31)

4.1.2 BMI and long-term outcome after TKA (Study II)

Between May 1997 and May 2000, a total of 95 knee arthroplasties were performed in the Department of Orthopaedics and Traumatology, Kuopio University Hospital, in 85 randomly selected patients who met the inclusion criteria: no previous knee or hip operations, no medication or diseases known to influence bone mineral metabolism. Two other TKA studies were conducted at the same time and those patients were not included into this study. After obtaining the approval of the ethics committee of the University of the Kuopio, the current study continued until March 2009 by inviting all living patients to a follow-up examination. Eighteen patients had died (1997-2009) during the follow-up and in addition, 27 patients were not able to participate because of miscellaneous reasons (unfit n = 8, change of residence n = 2, cancelled without stating reason n = 17). The final follow-up cohort consisted of 48 TKA patients and 52 knees. The minimum duration of the follow-up period was nine years (mean 10.8 years, range 9-12 years). The mean age of the patients at the time of follow-up examination (April 2009) was 76.3 years (54−90, SD 6.7).

Four knee implant designs from four manufacturers were used: AMK (n = 8) (DePuy, Warsaw, IN, USA), AGC (n = 2) (Biomet Merck, Limited, Bridgend, South Wales, UK), Duracon (n = 25) (Stryker, Howmedica, NJ, US) and Nexgen (n = 17) (Zimmer, Warsaw, IN, US). A posterior cruciate retaining technique was performed with 46 knees and both anterior and posterior cruciate ligaments were removed in six cases.

The clinical and radiological evaluations were performed preoperatively, and at the follow-up visits (1-year postoperatively and April 2009). The clinical evaluations included the Knee Society clinical rating system (KSS) which contains two components of assessments (Knee score (KS) for assessing the condition of the knee and the prostheses and Function score (FS) for measuring the patient’s functional abilities) with a maximum of 100 score for each parameter (Insall et al. 1989). The knee score evaluates pain, range of motion (ROM) and knee stability, the function score evaluates walking distance, ability to climb stairs and the need for ambulatory supports.

The clinical evaluation also included Western Ontario and McMaster Universities Osteoarthritis index (WOMAC), and the Timed- Up-and Go-test. An extensive medical examination and interview were also performed.

There were 43 knees with preoperative varus deformity (mean 10.3°, SD 5.4) and 9 knees with preoperative valgus deformity (mean 7.7°, SD 4.5). The mean age of the patients at the time of the follow-up examination (April 2009) was 76.3 years (54−90, SD 6.7).

The patients were divided into two study groups according to the value of their body mass index (BMI). BMI was calculated by dividing the subject’s weight in kilograms by their height in meters squared (kg/m²). BMI was categorized as non-obese (Group 1) and obese (Group 2). Group 1 consisted of patients whose body mass index was < 30 kg/m ² (non-obese, n = 26 preop/1-year; n = 23 spring 2009) and the patients with BMI ≥ 30 kg/m² were designated into Group 2 (obese, n = 22 preop/1-year; n = 25 spring 2009). A total of five patients in the primary group 1 had gained and two patients in the primary group 2 had reduced their weight over the cut-off line during the mean of 10.8 years’ follow-up.

(32)

4.1.3 Primary TKA and UKA revision patients (Study III)

During 1991-2000, a total of 2949 knee, 344 unicompartmental knee and 484 revision knee arthroplasties were performed in the Department of Orthopaedics and Traumatology, Kuopio University Hospital. During this time period, 44 revisions needed to be performed on failed medial unicompartmental knee arthroplasties. Primary UKA operations were performed with Duracon, Kirschner, Link Endomodel, Miller-Galante and PCA semiendoprosthesis. All of these UKA revision patients included in this study had primarily medial unicompartmental knee osteoarthritis. The decision to perform the primary UKA was made by the surgeon during the operation. After approval by the ethics committee of Kuopio University, the project started in May 2008 by sending the patients a letter of invitation to attend a control examination in Kuopio University Hospital. Sixteen UKA revised patients had died during this time period (1991−2008) and six UKA revised patients, because of different reasons (dysphoria n = 2, cancelled without reporting the reason n =2, excellent knee n = 2), were not able to participate in this study. Furthermore, one patient did not fit the inclusion criteria because her UKA had been converted to a new UKA. Thus a total of 21 of the original 44 UKA revision patients were available with a minimum of 8 years follow-up (mean, 10.5 years; range 8−17 years) and are included in this report.

In many studies, the outcome after UKA revision operations has been reported to be comparable to that achieved in primary TKA operations (Levine et al. 1996, Chakrabarty et al. 1998, Foong et al. 2014). In this present study, the UKA revision patients (Group 1) were compared to a randomly selected group of TKA patients (n = 56) (Group 2) whose age, sex and the operation time point were comparable to those in Group 1. The number of control patients was based on the living UKA revision cases multiplied by 2. Patients for the control group were randomly selected from the population of TKA patients having the same age, gender and operation time distributions as UKA revised patients. Sixteen patients had died in this group and 12 patients were not able to participate in this study for miscellaneous reasons (dysphoria n = 4, cancelled without reporting a reason n = 5, bed patient n = 1, not able to communicate n = 1, moved n = 1). Thus TKA group (Group 2) consisted of 28 patients who had undergone primary TKA operation at the same time as the UKA patients had undergone the revision operation. Hence the final study population consisted of 49 patients (Group 1, n = 21, Group 2, n = 28).

The time from primary UKA to UKA revision as well as the mode of failure and requirement of augments, stems and bone grafts was recorded. Four knee implant designs from three manufacturers had been used for revision operation: AMK (1) (Depuy, Warsaw, IN, USA), Duracon (6) (Stryker Howmedica, NJ, USA), Miller-Galante (3) (Zimmer, Warsaw, IN, USA) and NexGen (11) (Zimmer, Warsaw, IN, USA). The first follow-up visit was scheduled at approximately 3 months after the operation for patients in both groups. The postoperative 3-month range of motion (ROM) and postoperative complications were also recorded for the first time at that visit.

The patients were invited to come to the Kuopio University Hospital for follow-up control at June 2008 and a detailed clinical examination and interview were performed.

The knee status included: ROM, stability, pain, cicatrix length and self-reported walking

(33)

distance. Walking ability and balance were measured with timed Up- and Go-test. The present BMI was measured and obesity was defined as BMI greater than 30 kg/m2. The current need for an ambulatory support was also recorded. Each patient completed also the WOMAC- questionnaire.

The mean age of the patients at the time of the follow-up control (June 2008) was 75 years (SD, 7.21; range, 61−87). There were 20 men and 29 women. The mean body mass index was 30.1 kg/m² (SD, 5.67; range, 18.3−47.0). A total of 23 patients were considered obese based on a BMI value greater than 30 kg/m², 8 of those were men and 15 women (p

= 0.292). There were no statistically significant differences between the study groups (Group 1 = UKA revision, Group 2 = TKA primary) with regard to the measured characteristics, with the exception of their smoking habits. Demographic values and descriptive statistics are shown in Table 3. The postoperative hospital stay for the patients who underwent UKA revision averaged 8.3 days (SD 1.05, range 8−11) and for the patients who underwent primary TKA 7.9 days (SD 1.83, range 6−12).

Table 3. Demographic data of the 49 patients in study III (SD).

Group 1 (n = 21) (UKA rev)

Group 2 (n = 28) (TKA prim)

p-value Male/Female (N) 9/12 11/17 0.80

Age (years) 74.9 (7.4) 75.2 (7.2) 0.88

Chronic diseases (N, yes) Diabetes

ASO

Hypertension Cardiac

17 6 2 16 8

25 11 0 19 15

0.41 0.44 0.095 0.52 0.28

Smokers (N, yes) 4 0 0.016

BMI kg/m²(follow-up visit) 28.6 (4.59) 31.3 (6.22) 0.23 BMI kg/m² (perioperative) 28.5 (3.99) 30.5 (4.43) 0.28

4.1.4 DXA follow-up (Study IV)

In study IV, the subjects partially consisted of the same population as in study II. The fourth study focused on assessing the long-term changes in the bone mineral density of the distal femur between obese and non-obese patients. All the operations and DXA-based follow-up measurements were carried out between 1997 and 2007. Unfortunately, not all patients were able to participate in each of the DXA measurements which led to a loss of the study population during follow-up. Three patients had died and furthermore some had a poor health status or expressed a subjective desire to quit. Therefore, 61 patients (with 69 knees) were able to participate in the final DXA investigation at a mean of seven years after the operation (Duracon n = 34, Nexgen n = 21, AMK n = 11, AGC n = 3).

(34)

BMD (g/cm2) of the distal femur was measured at baseline, 3- and 6- months postoperatively and 1-, 2-, 4- and 7-years follow-up by using a fan-beam dual X-ray absorptiometry (Expert XL, Lunar Co., Madison, WI). The regions of interest (ROIs 1-5) were both metaphyseal and diaphyseal (Fig. 2).

The clinical evaluation of the patients was performed preoperatively and at each postoperative follow-up visit. The clinical evaluation included the Knee Society clinical rating system evaluation (KSS) which contains two different components: Knee Score (KS) and Function Score (FS) with a maximum value of 100 points for each component. A detailed medical examination and interview were also conducted at each visit.

The patients were divided into two study groups according to BMI. Group 1, non-obese (n = 35 patients/ 39 knees), consisted of patients whose body mass index was < 30 kg/m2.

Group 2, obese (n = 26 patients/ 30 knees), consisted of patients whose BMI was ≥ 30 kg/m2. The basic demographic data of the patients in study IV is expressed in table 4.

Table 4. Demographic data of 61 OA patients in study IV.

Description Number/mean

Male/Female Age (years,SD) BMI (kg/m²) Number of Obese Preop. KS

Preop. FS

12/49 67.7 (6.2) 29.8 (5.0) 26

51.5 (14.8) 56.6 (21.3)

Bone mineral density (g/cm2) was analyzed using the software algorithm provided by the manufacturer. This enables the quantization of bone mineral density in the presence of metal in the scanning field. In order to minimize operator-related inaccuracies associated with manual drawing, no attempt was made to exclude the cement mantel from the analysis. During the lateral scan, the patient lay on his/her side with the knee flexed at an angle of 15 degrees and two repeated scans were performed. The prostheses flanges were used for determining the location of regions of interest (ROI).

After locating the ROIs manually for the first scan, the ‛compare facility ‛ of the software was used for copying the ROIs for the second and repeated scans during follow- up. Apart from using Lunar manuals, also a personal communication from the Lunar Company was provided to help with the edge-detection algorithms and recommendations on the exploitation of plastic backgrounds for better and more reliable definition of bone, soft tissue and air outlines (Soininvaara 2005).

(35)

Figure 2. Periprosthetic regions of interest (ROIs) of the distal femur: ROI1 Anterior metaphyseal, ROI2 Central, ROI3 Posterior, ROI4 Diaphyseal and ROI5 Total metaphyseal, marked with arrow = ROIs 1+2+3.

4.3 STATISTICAL ANALYSIS

Statistical tests were performed with SPSS (SPSS Inc., Chicago, Ill., USA) software. The level of significance was set to p-value ≤ 0.05 in all statistical analysis in this thesis and standard descriptive statistics were calculated for the study population. Student’s T-test was used to assess group differences in continuous variables and Mann-Whitney test to evaluate group differences in continuous variables with non-normal distributions.

Wilcoxon’s signed-rank test was used for the case of repeated measurements and the analysis for categorical parameters was performed with Pearson’s chi-square test in studies I, II and III. The correlations between patients’ BMI or age and continuous variables in the studies I and III were assessed by Spearman’s non-parametric correlation matrix. In study IV, the trends of the BMD changes of the whole study cohort were performed with Greenhouse-Gaisser epsilon corrected repeated measures ANOVA (General linear model). DXA measurement data was noted to be normally distributed by the Kolmogorov-Smirnov-test and by examining the dispersion patterns.

Patient and operation related factors influencing the outcome of knee arthroplasty

uef.fi

Dissertations in Health Sciences

JAAKKO JÄRVENPÄÄ

PATIENT AND OPERATION RELATED FACTORS INFLUENCING THE OUTCOME OF KNEE ARTHROPLASTY

JAAKKO JÄRVENPÄÄ

Patient and operation related factors

influencing the outcome of knee arthroplasty

To Sanna, Viljami and Helmi

Acknowledgements

List of the original publications

Contents

Abbreviations

1 Introduction

2 Review of the literature

3 Aims of the study

4 Materials and methods