Effect of Body Mass Index on the outcomes of primary Total Knee Arthroplasty up to one year – A prospective study

Abstract

Background

Obesity is a significant risk factor for developing knee osteoarthritis, and these cases suffer from complications following Total Knee Arthroplasty (TKA). An association between obesity and outcome after TKA is ambiguous. Knowledge is scarce about a definite relation between the two. This study aims to establish a correlation between obesity and early outcomes of TKA.

Methods

This prospective cross-sectional study was done in cases undergoing primary knee arthroplasty between September 2019 to August 2020. Obesity was classified in all cases, and multiple variables like pain, functional status, Range of Motion, knee deformity, and Patient Response Outcome Measures were recorded. Statistical analysis was performed using SPSS Statistical Software version 22.0 and R.3.2.0. The level of statistical significance was taken as p < 0.05.

Results

We studied 100 knees (37 bilateral and 26 unilateral) in 63 cases. Pain score decreased maximally in the normal and overweight group and minimal in class III obesity (p < 0.001). KSS, FKSS, and PROMs gradually improved in all, except in morbidly obese (p < 0.001). Although the improvement in all variables was minimum in class III obesity compared to other classes of obesity, the margin of difference from the preoperative period was maximum in class III obese participants.

Conclusion

All cases, irrespective of class of obesity, experienced a comparable improvement in their knee function and improved quality of life. In addition, the TKA offered substantial benefits in terms of pain relief, knee stability, walking distance, range of movement of the knee, and stair climbing.

1. Introduction

According to World Health Organization (WHO), obesity is a medical condition that can be quantified by the Body Mass Index (BMI) and is closely related to total body fat percentages. WHO has classified obesity according to the BMI into four grades- Grade I- Overweight (BMI- 25.0-29.9), Grade II- Class I Obesity (BMI- 30-34.9), Grade III- Class II Obesity (BMI-35.0-39.9), and Grade IV- Class III Obesity (BMI >40).¹

Obesity is considered a significant risk factor for developing knee osteoarthritis (OA), with nearly a three-to five-fold increased risk compared with patients who are not overweight.^{2, 3, 4} Obese individuals are more likely to develop knee OA earlier in life and suffer from complications following Total Knee Arthroplasty (TKA).^5,6 Risk of developing OA increases by 36% for every 5-kg increase in BMI.⁷ Although TKA is the most effective treatment for patients with severe knee OA,⁸ obese patients with advanced knee OA are considered poor candidates for TKA. A high BMI is associated with limited physical performance and functional mobility in elderly persons or patients with OA and is associated with post-TKA physical disability.⁹ The BMI of cases requiring TKA has increased over time, and the relationship between high BMI and the risk of subsequent TKA has become a focal point of research.¹⁰

The correlation between knee OA and obesity is well recognized for several years. However, the association between obesity and outcome after TKA is ambiguous. There is still a scarcity of knowledge in the literature about a definite relation between the two. This study aims to establish a correlation between obesity and early outcomes of TKA.

1.1. Inclusion criteria

• •Patients with severe (Grade IV) Osteoarthritis or
• •Patients with moderate (Grade III) Osteoarthritis with gross functional limitation
• •Patients who gave Informed Consent

1.2. Exclusion criteria

• •Patients of rheumatoid arthritis undergoing TKA
• •Patients who underwent previous uni-condylar knee replacement
• •Patients who underwent Previous High Tibial Osteotomy (HTO).
• •Patients having Haemophilic knee joint arthritis
• •Patients with Gouty arthritis
• •Patients having damage to the knee joint attributed to vascular aetiology
• •Patients with post-traumatic knee OA, including previous fractures or dislocation, knee instability, and post-menisectomy

2. Material and methods

After taking an institutional ethical clearance, a prospective, observational, cross-sectional study was done on patients undergoing primary TKA for one year (September 2019–August 2020) at a tertiary care hospital. We included all moderate to severe knee OA cases undergoing TKA. The knees with inflammatory arthritis and any prior surgical interventions (e.g., arthroscopic surgery, corrective osteotomy, etc.) were excluded.

All the participants were classified as per their BMI (according to WHO classification), and multiple variables like pain, functional status, Range of Motion (ROM), Deformity of the knee, and Patient Response Outcome Measures (PROMs) were recorded pre-operatively. In addition, the pain was assessed by the Visual Analog Scale (VAS), Functional status was assessed by Knee Society Score (KSS), and Functional Knee Society Score (FKSS). All the implants used were standard and surgeries performed were posterior cruciate sacrificing (PS) type TKR.

The postoperative assessment was done for the pre-operatively recorded variables and postoperative complications in four-time periods- A) Immediate (Day1-Day 10), B) intermediate (Day10–6 weeks), C) late (6 weeks to 6months), and D) final (6months-1 year) following the index surgery.

Statistical analysis was performed using SPSS Statistical Software version 22.0 and R.3.2.0. Clinical Parameters were presented in Mean and Standard Deviation (SD) for quantitative variables and frequency (%) for qualitative variables. Wilcoxon Rank-Sum Test was applied to compare the clinical parameters like Blood loss, Length of Stay between groups of patients (Group-1: BMI<30, Group-2: BMI>30). Wilcoxon Signed-Rank test was used to compare parameters at different follow-ups. The level of statistical significance was taken as p < 0.05.

Variables and frequency (%) for qualitative variables. Wilcoxon Rank-Sum Test was applied to compare the clinical parameters like Blood loss, Length of Stay between groups of patients (Group-1: BMI<30, Group-2: BMI>30). Wilcoxon Signed-Rank test was used to compare parameters at different follow-ups. The level of statistical significance was taken as p < 0.05.

3. Results

A total of 152 knees in 100 patients were initially screened for this study. We excluded 37 patients (52 knees) who did not meet the inclusion criteria. We thus studied 63 patients with 100 knees (37 bilateral and 26 unilateral) in this study. Among 100 knees, distribution in various categories of obesity was normal (4), overweight (18), class I obesity (30), class II obesity (46), and class III obesity (2).

A transition to a higher class of obesity was associated with poorer outcomes. The pain score gradually decreased with time in all the classes of obesity (Table 1), with a maximum decrease in a normal and overweight and a minimum decrease in class III obesity (p < 0.001). KSS and FKSS gradually improved with time in all classes (Table 1), with morbidly obese patients having a minor improvement (p < 0.001). PROMs were better in all classes of obesity (Table 2), as compared to the preoperative condition and class III obesity had the least satisfactory PROMs (p < 0.001).

Table 1.

Comparison of different classes of obesity with pain score, knee society score and patient function score.

PAIN SCORE	Class 1	Class 2	Class 3	Normal	Overweight	p-value
Pre-op	7.7	7.9	9	6.5	7.2	<0.001
Immediate	4.2	3.8	6	3	4	<0.001
Intermediate	2.4	2.0	4	2	2.3	<0.001
Late	0.8	2	3	0	1.1	<0.001
Final	0	0	2	0	0	<0.001
KNEE SOCIETY SCOREx
Pre-op	47.7	49.2	43	49.2	44.5	<0.001
Immediate	70.8	71.5	68	72	68.1	<0.001
Intermediate	72.7	72.7	68.5	77.5	73.2	<0.001
Late	82.8	82.8	72	83	80.8	<0.001
Final	88.6	88.4	82	88.7	87.7	<0.001
PATIENT FUNCTION SCORE
Pre-op	24.5	28.2	55	55	31.9	<0.001
Immediate	40	30	25	25	30	<0.001
Intermediate	72.8	55	55	55	55	<0.001
Late	77.3	72.9	75	75	71.6	<0.001
Final	80	80	80	80	80	<0.001

Table 2.

Comparison of different classes of obesity with patient response outcomes.

PATIENT RESPONSE OUTCOME	Class 1	Class 2	Class 3	Normal	Overweight	p-value
Pre-op	52.1	53.5	59	54	52.2	<0.001
Late	27	27.4	31	24	22.7	<0.001
Final	19.8	20	23	20	19	<0.001

With the increase in the class of obesity, the time taken for mobilization and length of stay in hospital was prolonged, blood loss and need for transfusion increased. On the other hand, time taken to return to activity of daily life and ROM to the normal range was maximum in class III obesity (Table 3), [p < 0.001].

Table 3.

Comparison of different classes of obesity with immediate and intermediate outcomes.

IMMEDIATE (post op day 1- day10)	Class 1	Class 2		Class 3		Normal		Overweight		p-value
Mobilization (in days)	1.5	2		2.5		1		1.11		<0.001
Length of hospital stay (in days)	4	4		5		4		4.11		<0.001
Blood loss-intra op (in ml)	255.8	273.1		350		222.5		240		<0.001
Blood loss-post op (in ml)	157.5	177.3		285		85		117.8		<0.001
INTERMEDIATE (post op day 11–6 weeks)
Return to daily activity (in weeks)	4	4.04	6		3		3.72		<0.001
ROM to normal (in weeks)	4	4	6		3.2		4		<0.001

During the final follow-up, all classes of obesity had similar functional outcomes with no residual deformity, and PROMs also were identical (Table 4), [p < 0.001]. Although the improvement in all variables was minimum in class III obesity compared to other classes of obesity, the margin of difference from the preoperative period was maximum in class III obese participants. Patient satisfaction after surgery and improvement in quality of life was also maximum in class III.

Table 4.

Comparison of different classes of obesity with late and final outcomes.

LATE (6 weeks–6 months)	Class 1	Class 2	Class 3	Normal	Overweight	p-value
Functional outcomes	77.3	72.9	75	75	71.6	<0.001
PROMS	27	23	27	24	22.7	<0.001
FINAL FOLLOW-UP (1 year)
Functional outcomes	80	80	80	80	80	<0.001
PROMS	19.8	20	23	20	19	<0.001

4. Discussion

The association between BMI and outcomes of TKA is unclear. Some studies have shown an increase in BMI had a negative effect on the results, whereas the others showed no relationship between BMI and the outcomes. Pasquier et al. reported varus deformity and morbid obesity were associated with a greater gain in flexion postoperatively, although, the final flexion was lower than the mean in the overall population.¹¹ We noticed that the difference between preoperative and final flexion achieved was maximum in class III obesity (Table 1).

Obese patients undergoing TKR should be specifically informed regarding potential additional risks of complications and poorer outcomes compared to normal or overweight patients.¹² Effect of significant weight loss in morbidly obese patients by lifestyle modification and/or bariatric surgery and its implications on ligament imbalance, early failure due to aseptic loosening due to asymmetrical forces on the prosthesis needs to be explored. Various variables determine the outcomes following TKR like pre-operative functional status, sarcopenia, osteopenia, patient mental status, social support during rehab, associated comorbidities, and all of these should be evaluated before considering a patient for TKR.

Singh et al. found that BMI had no effect on postoperative joint perception, and all classes of obesity experienced similar functional outcomes following TKA.¹³ The difference in the functional outcome in the immediate, intermediate, and late postoperative period but the final functional outcome after one year from the surgery was noticed in our study, in all obesity classes (Table 4 and Fig. 1). The cases with a higher BMI are at higher risk of lower functional scores and developing complications following TKA.¹⁴ Similarly, the maximum improvement in functional scores with the least incidence of infection was seen in a normal and overweight group. In contrast, minimum improvement in functional scores with the highest incidence of infection was seen in the morbidly obese group. The trend of decrease in the final functional score was seen while going up in the categories of obesity (Table 1, Table 3).

Fig. 1.

Comparison of classes of obesity with the final outcomes of TKA.

It is documented that the obese patient had an increased risk of medical and surgical complications with an increase in rates of complications with an increase in the class of obesity.¹⁵ We observed that all the perioperative complications were maximum in the morbidly obese group and were minimum in the normal or overweight group. The risk of developing complications increased with a transition to higher classes of obesity (Table 3).

An adverse effect of obesity on implant survival is often debated. Gaillard et al. reported that obesity did not affect mid-term implant survival, though their results showed poor functional outcomes and risk of postoperative complications in obese patients.¹⁶ We also observed that the implant positioning, radio-lucent lines around the implants, and implant survival at one year did not correlate with the grades of BMI or classes of obesity. It is a common belief that overloading of the knee occurs with high BMI causing more significant impact loading across the tibial component resulting in poor implant survival.^{17, 18, 19} But, it is found that a more sedentary lifestyle in morbidly obese patients counterbalances implant wear.

Our mean preoperative and postoperative overall KSS and FKSS were consistently lower in Class III obesity, although the difference between the two was maximum in class III obesity. Chen et al. studied 117 morbidly obese and 2108 non-obese patients and found comparable FKSS between the two groups and superior improvement in KSS in obese patients.²⁰ However, our final KSS and the FKSS were similar in all classes of obesity. Krushell et al. reported the minimum mean improvement in FKSS of 13 in the morbidly obese group (class III obesity) vs. 26 in the non-obese group.²¹ Their results signify that a morbidly obese patient who was housebound earlier can now walk for a significant distance postoperatively.

We were lucky not to encounter any postoperative infections in our subjects in the immediate postoperative period and up to the first year of their follow-up. Still, these patients need to be observed closely in the future for any delayed complications like prosthetic joint infections and implant loosening. We suggest that the morbid obesity group should not be excluded from TKA based on their BMI alone, as this procedure significantly improves functional capability and quality of life in these individuals.²²

5. Limitations

We acknowledge some limitations of this study viz., the sample size was not large, and it was a single-center study. Although the follow-up period was up to one year, this study describes the early clinical outcomes in obese that are considered high-risk individuals for TKA.

6. Conclusion

All cases, irrespective of class of obesity, experienced a comparable improvement in their knee function and improved quality of life. Also, TKA in our cohort offered substantial benefits in pain relief, knee stability, walking distance, range of movement of the knee, and stair climbing.

Declaration of competing interest

None.