Frailty and sarcopenia as predictors of postoperative outcomes in elderly patients undergoing lower limb orthopedic surgeries in South India

Abstract

Background and Aims:

As the aging Indian population expands, anesthesiologists will increasingly encounter frail surgical patients. We aimed to study the prevalence of frailty, its association with sarcopenia, and its impact on postoperative outcomes among older patients (>65 years) undergoing lower limb orthopedic surgeries.

Material and Methods:

Frailty was assessed using the Clinical Frailty Score (CFS), and values were compared with ultrasound-guided quadriceps muscle thickness measurements for sarcopenia, after normalizing for body mass index (BMI) and body surface area (BSA). A comparison of perioperative characteristics and postoperative outcomes with frailty was made using Spearman’s correlation and the Chi-square test. An ROC curve was plotted to evaluate the validity of the normalized quadriceps muscle thickness value in diagnosing frailty.

Results:

The proportion of frailty in our study population was 80.1% with a significant association for sarcopenia (Spearman’s correlation coefficient -0.182 and -0.237 for quadriceps muscle thickness normalized for BMI and BSA, P value < 0.001), age (72.35 (64.2, 80.5) versus 68.76 (64.97,72.55) years, P = 0.001), ASA classification (P < 0.001), and surgical type (P = 0.006). Compared to the nonfrail, patients with frailty had higher rates of postoperative ICU admission (P = 0.003), increased lengths of ICU (P = 0.005) and hospital stay (P = 0.005), and more postoperative complications (39.4% vs 12.9% patients, (P = 0.001), OR 3.061 (1.636-5.730)).

Conclusion:

Frailty occurs in a large percentage of older orthopedic patients and significantly impacts postoperative outcomes.

Introduction

As the world population ages, the older Indian population is expected to reach 1.7 billion by the early 2060s.[1] Findings from the Longitudinal Aging Study in India report a frailty prevalence of 30% among adults over 45 years, with women being twice as likely to be frail compared to their male counterparts, and more vulnerable to falls, cognitive deficits, and hospitalization.[2] According to a recently published market survey, the number of elderly joint replacement surgeries in India is increasing annually, with hip arthroplasties predicted to grow at the highest rate globally, between 2020 and 2026.[3] Despite this, a significant knowledge gap exists regarding the proportion of frailty among elderly Indian surgical patients and its effects on postoperative outcomes.

This prospective, observational study aimed to determine the proportion of frailty in elderly adults (over 65 years) presenting for elective lower limb orthopedic surgeries at a tertiary care center in South India. The secondary objectives were to correlate quadriceps muscle sarcopenia (as assessed by point-of-care bedside ultrasonography) with frailty and to study its association with intraoperative and postoperative complications, duration of hospital stay, and overall outcomes (i.e., death or discharge from the hospital). We hypothesized that frailty, as assessed by the Clinical Frailty Score, would be associated with reduced ultrasound-guided quadriceps muscle thickness (after normalizing for BMI and BSA) and linked to poorer perioperative and postoperative outcomes.

Material and Methods

This prospective observational study was conducted in a tertiary care referral hospital in South India after obtaining Institutional Research and Ethics Committee clearance (BCMCH/IEC/2022/09/300) and CTRI registration (CTRI/2023/11/060237). Written informed consent was obtained from all patients for participation in the study, including the use of patient data for research and educational purposes. The study procedures adhered to the principles of the Declaration of Helsinki, 2013, and good clinical practice.

The inclusion criteria were patients over 65 years of age, belonging to ASA physical status classifications I, II, or III, and undergoing elective lower limb orthopedic surgeries. Patients who refused consent; those with extremity amputations, paraplegia, or known neuromuscular diseases; and those undergoing emergency procedures were excluded.

The sample size was calculated using OpenEpi version 3.01 (available at https://www.openepi.com/SampleSize/SSPropor.htm). Considering the prevalence of frailty among nononcologic surgical patients as 30–40%,[4] with 80% power, a 95% confidence level, and absolute precision (d) of 5%, the calculated sample size was 336, rounded off to 350 patients. We expected to reach our recruitment target within 15 months from the start of the study.

After obtaining informed consent, the following demographic variables were collected: name, age, sex, type and date of surgery, height, weight, ASA classification, and the presence of any comorbidities. Frailty was assessed during the preanesthetic visit before surgery using the Clinical Frailty Score (CFS), which classifies patients on a 9-point ordinal scale. A score of ≥5 was considered frail (CFS 5, 6, and ≥7 were classified as mild, moderate, and severe frailty, respectively).

On the morning of surgery, an ultrasonic assessment of the quadriceps muscle thickness was performed using the Sonosite M-Turbo ultrasound system (Fujifilm Sonosite, USA) with a linear-array transducer (13–6 MHz bandwidth). Two-dimensional ultrasound images were obtained with the patient supine, at 30° upper body elevation, and legs extended. Measurements were taken on the nonoperative leg (the right leg in cases of bilateral limb surgeries) at the 60% length mark measured from the anterior superior iliac spine to the superior border of the patella. Three measurements were taken, and the average was computed. All ultrasound assessments were performed by one of two Anesthesiology residents who had completed a 2-day hands-on training session led by a radiology consultant specifically for this study. Interobserver variability was minimized by standardizing patient positioning and anatomical landmarks for the single, linear (depth) measurement.

Intraoperatively, data were recorded on the type of anesthesia administered, the occurrence of intraoperative complications (classified under the ClassIntra score), and the location of postoperative transfer (ICU, HDU, Recovery Room, or Ward). All patients were followed up daily, and outcomes were recorded, including the presence of postoperative complications (using the Clavien–Dindo classification), duration of hospital and intensive care unit stay, and overall outcomes (death or discharge). The study was completed within 12 months (January to December 2024).

Statistical analysis was performed using SPSS version 21.0 (IBM Corp., Released 2012. IBM SPSS Statistics for Windows, Version 21.0. Armonk, NY: IBM Corp.). Continuous and categorical variables were presented as mean/median and counts/percentages, respectively. To elucidate the association of frailty with perioperative variables (age, surgical procedure, comorbidities, intraoperative and postoperative complications, location of postoperative transfer, duration of ICU and hospital stay, and death) and normalized quadriceps thickness values, the Chi-square test and Spearman’s correlation were used. The body mass index (BMI) and body surface area (BSA) of the study patients followed a normal distribution as per the Kolmogorov–Smirnov test. Quadriceps muscle thickness values were normalized for BMI and BSA to account for variations due to age, sex, body stature, height, and ethnicity among patients. An ROC curve was plotted using the normalized quadriceps thickness values against frailty, and the area under the curve (AUC) was estimated. For all statistical evaluations, a P value of less than 0.05 (P < 0.05) was considered significant.

Results

A total of 524 patients undergoing lower limb orthopedic surgeries were screened, of which 361 met the inclusion and exclusion criteria for the study. Complete data and follow-up notes were available in 352 patients and included for final analysis. The proportion of frailty among our study population using the CFS was 80.1% (282 out of 352 patients) with mild frailty in 109 patients (31%), moderate in 100 patients (28.4%), and severe frailty in 73 patients (20.8%). Frailty was found to be significantly related to age, 72.35 (64.2, 80.5) versus 68.76 (64.97,72.55) years, P = 0.001, ASA physical status classification (P = 0.000), and in those undergoing lower limb orthopedic surgery (P = 0.006). Among the various comorbidities studied, patients with associated obstructive airway disease were more likely to be frail (P = 0.049). Other demographic data and baseline characteristics showed no statistical differences between the two groups [Table 1].

Table 1.

Baseline characteristics and comorbidities of patients undergoing elective lower limb orthopedic surgeries, by frailty status

Variable	Frail (n=282 patients)	Nonfrail (n=70 patients)	P
Age, mean±SD	72.35years (±8.15)	68.76years (±3.79)	0.001
Gender- Female (277 patients)	222 (80.1%)	55 (19.9%)	1.000
BMI* kg/m2	28.09 (±5.19)	28.56 (±4.44)	0.079
BSA†	1.63(±0.18)	1.66 (±0.17)	0.522
ASA‡ physical status classification
• I (7 patients)	3 (42.9%)	4 (57.1%)	0.000
• II (199 patients)	142 (71.4%)	57 (28.6%)
• III (146 patients)	137 (93.8%)	9 (6.2%)
Comorbidities:
• Hypertension (249 patients)	200 (80.3%)	49 (19.7%)	0.879
• Diabetes Mellitus (139 patients)	115 (82.7%)	24 (17.3%)	0.320
• Dyslipidemia (97 patients)	74 (76.3%)	23 (23.7%)	0.267
• Coronary Artery Disease (41 patients)	37 (90.2%)	4 (9.8%)	0.097
• Obstructive Airway Disease (16 patients)	16 (100%)	0 (0%)	0.049
• Cerebrovascular disease (14 patients)	14 (100%)	0 (0%)	0.081
• Chronic Kidney Disease (14 patients)	13 (92.9%)	1 (7.1%)	0.318
• Thyroid disease (44 patients)	38 (81.8%)	8 (18.2%)	0.762
• Neurologic disease (18 patients)	17(94.4%)	1 (5.6%)	0.140
• Psychiatric disease (6 patients)	6 (100%)	0 (0%)	0.603
Type of Surgery:
• Knee Arthroplasty Unilateral (n=82 patients)	60 (73.2%)	22 (26.8%)	0.006
• Knee Arthroplasty – Bilateral (n=152 patients)	114 (75%)	38 (25%)
• Hip Arthroplasty (n=71 patients)	67 (94.4%)	4 (5.6%)
• Others (n=47 patients)	41(87.23%)	6 (12.7%)

*BMI- Body Mass Index, ^†BSA- Body Surface Area, ^‡ASA- American Society of Anaesthesiologists

The average overall quadriceps muscle thickness in the study population was 1.067 (0.204, 1.93) cm, and quadriceps thickness values normalized for BMI and BSA averaged 0.650 (0.12,1.18) cm and 0.038 (0.006,0.07) cm, respectively. Frailty was significantly associated with sarcopenia as determined by quadriceps muscle thickness normalized for BMI and BSA (Spearman’s correlation coefficient -0.182 and -0.237, respectively, P value < 0.001 in both). Receiver operating characteristic (ROC) analysis for quadriceps muscle depth showed the following values: When normalized by body mass index, the AUC value was 0.643 (0.57, 0.71), and when normalized by body surface area, it was 0.653 (0.58, 0.72), suggesting limited clinical utility as a diagnostic test for frailty in our patients [Figure 1]. Figure 2 shows the relative quadriceps muscle mass differences (normalized for BMI and BSA) for frail and nonfrail patients [Figure 2].

Figure 1.

Receiver operating characteristic (ROC) analysis for quadriceps muscle thickness in patients, normalized for body mass index, BMI (green curve) and body surface area, BSA (blue curve)

Figure 2.

Box and whisker plot showing relative quadriceps muscle thickness in frail (Blue box) and not-frail patients (Orange box), normalized for body mass index, BMI (left) and body surface area, BSA (right)

There were no differences between frail and nonfrail patients for intraoperative blood loss (P = 0.448), type of anesthesia (P = 0.427), and intraoperative complications (P = 0.262). However, the presence of frailty was significantly associated with the need for postoperative ICU admission (11.7% vs 0 patients, P = 0.003), an increased length of ICU stay, that is, 0.30 (-0.645, 1.245) days vs 0 days, P = 0.005, and longer hospital stay, that is, 7.39 (4.67,10.11) days vs 6.56 (4.57,8.55) days, P = 0.005. Postoperative complications, in general, were higher in frail compared to nonfrail patients (39.4% patients vs 12.9% patients, P = 0.001), with an odds ratio of 3.061 (1.636, 5.730). Individual postoperative complications [Supplementary Table A] between groups were not studied as their numbers were too small to make a meaningful comparison. There were no mortality differences between the two groups (P = 1.000) [Table 2].

Supplementary Table A.

Details of Post-operative Complications in Patients with and without Frailty

	Frequency	Percentage (%)	Frail (n=282 pts)	Non-Frail (n=70 pts)
Nil	232	65.9	171	61
Anemia	49	13.92	46	3
Cardiac events	36	10.22	35	1
Electrolyte abnormality	3	.85	3	0
Delirium	12	3.4	11	1
Infection	7	1.9	4	3
Respiratory complications	4	0.11	4	0
Renal complications	13	3.6	12	1
Uncontrolled sugars	10	2.8	10	0
Redo surgical procedures	7	1.9	7	0
Others (deranged LFTs, unexplained hypotension, drug allergies)	8	2.2	8	0
Total Complication Events	149	42.32	140	9

(More than one complication occurred in some patients)

Table 2.

Perioperative characteristics, complication rates and use of resources, by frailty status

Perioperative Characteristics	Frail (n=282 patients)	Nonfrail (n=70 patients)	P
Type of Anesthesia, SAB*/GA†	273/9	69/1	0.427
Class Intra score for Intraoperative complications
• 0 (n = 163)	125 (93.25%)	38 (23.31%)	0.286
• 1 (n = 20 patients)	19 (95%)	1 (5%)
• 2 (n = 158 patients)	128 (81.01%)	30 (18.99%)
• 3 (n = 8 patients)	7 (87.5%)	1 (22.5%)
• 4 (n = 3 patients)	3 (100%)	0
Intraoperative blood loss (> 250 ml, n = 22 patients)	19 (86.4%)	3 (13.6%)	0.448
Clavien-Dindo Score for Postoperative complications
• 1 (n = 226 patients)	165 (73%)	61 (27%)	0.000
• 2 (n = 77 patients)	70 (90.9%)	7 (9.1%)
• 3 (n = 18 patients)	16 (88.89%)	2 (11.11%)
• 4 (n = 29 patients)	29 (100%)	0
• 5 (n = 2 patients)	2 (100%)	0
Postoperative transfer to ICU (n = 33 patients)	33 (100%)	0	0.003
Requirement of post-op ventilatory support (n = 7 patients)	7 (100%)	0	0.214
Length of ICU‡ stay	0.30 days (±0.945)	0.00 days (±0.000)	0.001
Length of Hospital stay	7.39 days (±2.728)	6.56 days (±1.997)	0.027
Death (n = 2)	2 (100%)	0	1.000

*SAB- Subarachnoid block, ^†GA- General anesthesia, ^‡ICU- Intensive Care Unit

Discussion

This study highlights that more than 80% of older patients requiring lower limb orthopedic surgeries in our population are frail, with a close correlation between frailty and quadriceps muscular sarcopenia. Moreover, these patients are at greater risk of requiring intensive care admission, experiencing postoperative complications, and having longer ICU and hospital stays. The use of validated tools is essential for identifying people with frailty,[4,5] but many are time-consuming and difficult to use in busy clinics or large-scale population research. We found that the Clinical Frailty Scale (CFS) was easy to use and administer in our high-volume preanesthetic assessment clinic and could readily categorize patients into mild, moderate, and severe frailty categories. A recent meta-analysis endorsed the use of the CFS tool and reported its validity in terms of accuracy and feasibility among 35 different frailty instruments compared across 45 studies.[6] Quick to administer and demonstrating good inter-rater reliability, even among untrained medical trainees,[7] the CFS has been endorsed by the British Orthopaedic Association for categorizing frail patients into specialized care pathways.[8]

Using the CFS cutoff of ≥5 to define frailty,[9] our observed frailty rate (80.1%) was much higher than that reported in other countries. We attribute this to two main factors: a higher baseline prevalence of frailty among older adults in our geographic region, and the specific surgical population studied. A large study conducted across high-, middle-, and low-income countries found that India had the highest percentage (55.5%) of frailty among older adults living in the community.[10] Similarly, another South Indian study reported a 62% prevalence of frailty among community-dwelling older adults (95% CI: 56.4–66.4).[11] Furthermore, the population we studied—elderly patients undergoing lower limb orthopedic surgeries—is particularly prone to frailty as degenerative musculoskeletal conditions such as rheumatoid arthritis, osteoporosis, and osteoarthritis, which increase the risk of falls and subsequent surgeries, are strongly associated with frailty.[12] Adding to this is the high burden of multimorbidity, malnutrition, and an ageing population in our district.[13] No Indian study to date has specifically examined frailty in this vulnerable surgical cohort, and our findings provide valuable insights. The only other Indian study using the CFS in the perioperative setting reported a frailty incidence of 30.9%, but it studied a different population (cardiac surgical patients) and used a higher CFS cutoff of six.[14] In our study, elderly age, the presence of comorbidities, higher ASA classification, and decreased muscle mass (sarcopenia) were significantly associated with frailty, consistent with prior reports.[2,3,4] A recent prospective, registry-embedded, community-based cohort study from India reported that frail patients had higher Charlson Comorbidity Index scores (OR: 1.73, 95% CI: 1.39–2.15) and greater rates of malnutrition (mild/moderate OR: 1.90, 95% CI: 1.29–2.80; severe OR: 4.76, 95% CI: 2.10–10.77) associated with sarcopenia.[15]

Several studies have confirmed the validity of radiological measurements (both ultrasound and CT imaging) to assess sarcopenia in various muscle groups, such as the rectus femoris, quadriceps, quadratus lumborum, and psoas muscles.[16,17] However, we found that quadriceps muscle thickness, despite being quick and easy to perform, showed limited diagnostic efficacy for detecting frailty in our patients (AUC 0.64). This may be because frailty is influenced not only by muscle mass but also by muscle quality (including fat infiltration, fiber type changes, and neuromuscular function), which muscle thickness alone cannot assess. Additionally, variability in measurement techniques, including operator skill, probe placement pressure, and confounding factors such as edema and hydration status, may have affected its reliability. Potential ways to improve its utility include combining it with muscle quality metrics, integrating biomarkers, and leveraging machine learning models and AI tools.

In our study, the rates of adverse intraoperative events were similar between frail and nonfrail patients. However, frail older patients were significantly more likely to require ICU admission and to have longer ICU and hospital stays. Similar outcomes have been reported in large multicenter studies worldwide.[18] A meta-analysis of 16 studies involving 683,487 patients (444,885 of whom were frail) found that frail individuals had an increased likelihood of experiencing complications (RR: 1.48, 95% CI: 1.35–1.61; P < 0.001), a higher risk of readmission (RR: 1.61, 95% CI: 1.44–1.80; P < 0.001), and a greater likelihood of discharge to skilled care facilities (RR: 2.15, 95% CI: 1.92–2.40; P < 0.001).[19] Similarly, in a study of 432,828 patients from Veterans Health Administration hospitals, frailty was linked to higher mortality at 1, 3, and 6 months, particularly after low- and intermediate-risk surgeries compared to high-risk surgeries.[20]

Despite its high prevalence, frailty remains poorly understood by many orthopedic surgeons.[21] Newer guidelines on the perioperative care of people living with frailty recommend that frailty should be documented at the time of surgical referral, during preoperative assessment, and at hospital admission using the CFS. Patients with CFS >5 should undergo a comprehensive geriatric assessment and cognitive function evaluation.[22] Anesthesiologists are uniquely positioned to perform complete frailty evaluations as part of the perioperative care pathway and to raise awareness of frailty’s impact among their colleagues.[23] Studies show that prehabilitation—whether on its own or as part of a structured program—can enhance functional capacity and improve resilience to postoperative complications by providing personalized preoperative exercise regimens, optimizing nutrition, and offering psychological interventions for elective surgery candidates.[24,25] Given the high frailty rates identified in our study, these are promising and urgently needed directions for future implementation research in our country.

The strengths of our study include its prospective design, its focus on a high-volume, single-specialty surgical population, and the additional preoperative assessment of sarcopenia in the Indian population. To the best of our knowledge, only two other Indian studies (both nonorthopedic) have examined perioperative frailty assessment,[14,26] and none have evaluated sarcopenia. Our study also emphasizes the need for broader research on frailty’s impact and re-examination of current national geriatric health policies, given its morbidity burden and impact on surgical outcomes. The limitations of our study include that the findings cannot be extrapolated to nonorthopedic surgical groups, and long-term patient outcomes were not studied. Additionally, ultrasonic assessments of quadriceps thickness were performed by anesthesiology trainee residents rather than certified sonologists, and errors related to operator experience cannot be ruled out.

Conclusion

Early identification of frailty should become an essential component of routine preanesthetic assessments for elderly patients undergoing lower limb orthopedic procedures, to appropriately counsel and predict the risk of postoperative complications in this vulnerable surgical population.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.