Purpose
The purpose of this study is to use a large national database to assess short-term adverse events following arthroscopic rotator cuff repair in patients 65 years and older.
Methods
The ACS NSQIP database was queried to identify patients that underwent arthroscopic rotator cuff repair between December 31, 2015, and January 1, 2017. Patients were split into two groups: 1) between 40-65 years old and 2) 65+ years old. Cases involving open rotator cuff repair, total shoulder arthroplasty, hemiarthroplasty, and emergency surgery were excluded. Exact matching was used to control for confounding variables, including sex, body mass index (BMI), American Society of Anesthesiologists (ASA) score, operative time, and several medical comorbidities. After matching, the incidence of several 30-day patient complication measures was compared between the groups. Binary logistic regression was used to identify covariates associated with various 30-day complications.
Results
A total of 17,880 patients were included in the study. 69.4% (n = 12,404) patients were between 40 and 65 years old and 30.6% (n = 5,476) patients were 65+ years old. After matching, 9,210 patients were included in the final analysis. After matching, patients 65 years and older were more likely to experience 30-day unplanned readmission (P = .035) and overall medical complications (P = .036). There were no significant differences in most 30-day complication measures, including mortality (P = .250), reoperation (P = .449), non-home discharge (P = .255), surgical complications (P = .146), and several medical complications, including myocardial infarction (P = .165), deep venous thromboembolism (P = .206), pulmonary embolism (P = .196), and cerebrovascular accident (P > .999) between the two age groups.
Conclusions
In this matched cohort study, patients 65 years and older experienced a higher rate of 30-day unplanned readmission and overall medical complications following elective arthroscopic rotator cuff repair relative to patients under 65. However, these older patients did not have significantly worse rates of other 30-day complication measures, including mortality, reoperation, return to the OR, and non-home discharge.
Introduction
A substantial transformation in the United States’ age structure is currently under way, with the 65 and older cohort of Americans expected to double to nearly 90 million people between 2010 and 2050.
1
The wide-ranging implications of this shift are not only expected to affect the overall population, size, and diversity of the country, but also demands on social programs, including Social Security and Medicare. Within the world of healthcare, elective orthopaedic procedures are strongly linked to increased life expectancy and will likely see the greatest increase in demand, with rising incidence of conditions like osteoarthritis.2
Given this increasing life expectancy, it is important for physicians performing elective orthopaedic procedures to understand the risks specific to this older population. This will enable providers to address quality-of-life issues in balance with potential underlying medical comorbidities in order to optimize short-term outcomes in this population.Degenerative rotator cuff tear (RCT) and repair represent increasingly important issues as the American population ages.
3
In addition to being painful, degenerative RCT also has a major impact on patients’ ability to perform activities of daily living.4
The pain and functional disability associated with degenerative RCT is not only associated with a marked decrease in quality of life,5
,6
but is also a strong predictor of depression, anxiety, and sleep disturbance.7
Such findings once again call attention to identifying appropriate surgical candidates and balancing the risk of poor short-term outcomes with the potential to eliminate pain and improve function with surgical intervention.Degenerative rotator cuff tear (RCT) is already a common pathology, particularly in the elderly, and RCT repair is increasingly performed even in historically contraindicated cases involving considerable fatty degeneration.
8
Like many elective procedures, the decision to pursue operative versus nonoperative management for rotator cuff pathology is inherently challenging. This is further complicated in patients with advanced age, who may benefit less from surgery, but have a greater comorbidity profile. While these decisions are multifactorial, one consideration is the risk of short-term complications and poor outcomes. One study suggested that elderly patients were not at increased risk of short-term complications if these patients met specific health criteria prior to surgery.9
In contrast, another recent study found that being 65 years and older was associated with 30-day postoperative complications following arthroscopic rotator cuff repair.10
When assessing the risk of short-term complications, it is important to consider confounding variables, particularly in elderly patients who may present with comorbidities and poorer baseline health status. The purpose of this study is to use a large national database to assess the short-term adverse events following arthroscopic rotator cuff repair in patients 65 years and older. We hypothesize that after matching, patients aged 65 years and older will have similar short-term complication rates following arthroscopic rotator cuff repair relative to adult patients between 40 and 65 years old.Materials and Methods
Overview
This is a retrospective cohort study. All data were obtained from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP). Studies using the ACS NSQIP database have been designated as exempt by Institutional Review Board oversight, given that this is a database with deidentified data. The ACS NSQIP is a surgical outcomes database with proven reliability that was created to measure and improve the quality of surgical care.
11
,12
The data provided to users are deidentified as defined by the Health Insurance Portability and Accountability Act (HIPAA). Any attempt by researchers to identify patients is strictly prohibited. In the year 2017, there were over 1 million surgical cases reported by over 700 participating hospitals in the ACS NSQIP initiative.13
Each participating institution has a designated clinical reviewer that has been trained to record relevant data for various surgical cases.13
Auditing procedures ensure the reliability of reported data.12
The ACS NSQIP database has been employed to study short-term outcomes and complication rates after a wide array of surgical procedures, including the field of orthopaedics.14
, 15
, 16
Search and Inclusion/Exclusion Criteria
The ACS NSQIP database was queried to identify patients undergoing arthroscopic rotator cuff repair at participating institutions between January 1, 2015, and December 31, 2017. The Current Procedural Terminology (CPT) code 29827 was used to identify all cases. Cases were split into two groups based upon patient age at the time of surgery: 1) patients between 40 and 65 years old and 2) patients 65 years and older. Patients with open rotator cuff repair (CPT 23410, 23412, 23420), total shoulder arthroplasty (CPT 23472), hemiarthroplasty (CPT 23470), emergent surgery, and incomplete data were excluded for purposes of this study.
Predictor Variables
As outlined in the user guide of the ACS NSQIP, the database provides over 150 variables, including several preoperative risk factors. For this study, the following variables were extracted and assessed for each patient: sex, age, body mass index (BMI), American Society of Anesthesiologists (ASA) classification, and several medical comorbidities: diabetes, smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure (CHF), hypertension, renal failure, dialysis, chronic steroid use, bleeding disorder, and transfusion within 48 hours. Additionally, operative time was included as a variable of interest.
Exact Matching
To adjust for differences in patient health and other potential confounding variables between the two age groups, exact matching was used to create a 1:1 pairing of patients between the two groups. Patients were matched based on several predictor variables: sex, body mass index (BMI), American Society of Anesthesiologists (ASA) classification, and medical comorbidities, such as diabetes, smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure (CHF), hypertension, renal failure, dialysis, chronic steroid use, bleeding disorder, transfusion within 48 hours, and operative time. Of note, body mass index (BMI) was matched as a categorical variable: underweight (below 18.5), normal (18.5-24.9), overweight (25.0-29.9), obese class I (30.0-34.9), obese class II (35.0-39.9), and obese class III (40.0+). Operative time was also matched as a categorical variable: 1) under 60 minutes, 2) between 60-89 minutes, and 3) over 90 minutes. Cases were matched 1:1 based on exact matching of all covariates—meaning that each case pair had identical values for each of the aforementioned covariables. For cases that produced more than one identical match, the pairs were chosen through randomization.
Complication Measures
The ACS NSQIP database provides several 30-day patient complication measures, as outlined in the User Guide for the Participant Use Data File (PUF), which contains deidentified aggregate patient-level data available to researchers. For this study, the 30-day complications of interest were: return to the operating room, readmission, reoperation, non-home discharge, mortality, surgical complications (superficial and deep surgical site infections, dehiscence, bleeding requiring transfusion), and medical complications (wound infection, pneumonia, reintubation, failure to wean intubation, pulmonary embolism, renal insufficiency, renal failure, urinary tract infection, cerebrovascular accident, cardiac arrest, myocardial infarction, deep venous thromboembolism, systemic sepsis, and septic shock). The rate of these complications were compared between the two age groups of interest, both prior to and after matching.
Statistical Analyses
Pearson’s chi-squared and Fisher’s exact tests were used to compare the prevalence of the aforementioned predictor variables and complication measures between the two groups, both prior to and after exact matching. Fisher’s exact tests were used to compare rare events—specifically, if there were less than 5 patients in a category. Otherwise, all categorical variables were compared using chi-squared tests. Any continuous variables were compared using unpaired t-tests. After exact matching, binary logistic regression was used to identify any covariates (e.g., sex, body mass index (BMI), American Society of Anesthesiologists (ASA) classification, and medical comorbidities—diabetes, smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure (CHF), hypertension, renal failure, dialysis, chronic steroid use, bleeding disorder, transfusion within 48 hours, and operative time) that were associated with each individual 30-day complication measure. Of note, age was also modeled as a continuous variable to investigate an association between increasing age and various 30-day complications after matching for potential confounding variables. The cutoff for statistical significance was defined as P < .05 for all analyses. All statistical tests were completed using the IBM SPSS version 24 (Armonk, NY: IBM Corp).
Results
17,780 patients were included in the final analysis. Prior to matching, there were many significant differences in the patient-specific variables between the two age groups. Generally, older patients were more likely to have a poorer overall health compared to younger patients. Specifically, patients 65 years and older were significantly more likely to present as ASA Class 3 and 4+, as well as with comorbidities of diabetes, COPD, congestive heart failure, hypertension, chronic steroid use, and bleeding disorder. Younger patients were more significantly more likely to present with Class II and III obesity and were also significantly more likely to present as smokers (Table 1).
Table 1Demographics, Patient Variables, and Comorbidity Burden of the Cohort Prior to Matching
| 40-65 Years Old [n = 12,404] | 65+ Years Old [n = 5,476] | P | |||
|---|---|---|---|---|---|
| Sex | <.001 | ||||
| Male | 7,342 | 59.2% | 2,989 | 54.6% | |
| Female | 5,062 | 40.8% | 2.487 | 45.4% | |
| BMI (kg/m2) | |||||
| Underweight | 112 | 0.9% | 55 | 1.0% | .516 |
| Normal | 1,771 | 14.3% | 943 | 17.2% | <.001 |
| Overweight | 4,188 | 33.8% | 2,153 | 39.3% | <.001 |
| Obese Class I | 3,415 | 27.5% | 1,433 | 26.2% | .059 |
| Obese Class II | 1,743 | 14.1% | 592 | 10.8% | <.001 |
| Obese Class III | 1,175 | 9.5% | 300 | 5.5% | <.001 |
| Comorbidities | |||||
| Diabetes | 1,969 | 15.9% | 1,136 | 20.7% | <.001 |
| Smoking | 2,230 | 18.0% | 405 | 7.4% | <.001 |
| COPD | 318 | 2.6% | 241 | 4.4% | <.001 |
| Ascites | 0 | 0.0% | 1 | 0.02% | .132 |
| Congestive heart failure | 11 | 0.09% | 14 | 0.3% | .006 |
| Hypertension | 5,061 | 40.8% | 3,419 | 62.4% | <.001 |
| Renal failure | 0 | 0.0% | 2 | 0.04% | .094 |
| Dialysis | 14 | 0.1% | 10 | 0.2% | .240 |
| Chronic steroid use | 241 | 1.9% | 135 | 2.5% | .025 |
| Bleeding disorder | 128 | 1.0% | 124 | 2.3% | <.001 |
| Transfusion within 48 hours | 1 | 0.01% | 1 | 0.02% | .519 |
| ASA Class | |||||
| Class 1 (No disturbance) | 1,254 | 10.1% | 160 | 2.9% | <.001 |
| Class 2 (Mild disturbance) | 7,526 | 60.7% | 2,745 | 50.1% | <.001 |
| Class 3 (Severe disturbance) | 3,530 | 28.5% | 2,479 | 45.3% | <.001 |
| Class 4+ (Life threatening) | 91 | 0.7% | 88 | 1.6% | <.001 |
| Operative time | |||||
| Under 60 minutes | 3,274 | 1,618 | <.001 | ||
| Between 60 and 89 minutes | 4,061 | 1,905 | .004 | ||
| 90+ minute | 5,069 | 1,953 | <.001 | ||
ASA, American Society of Anesthesiologists; BMI, body mass index; COPD, chronic obstructive pulmonary disease.
Prior to matching, older patients had higher rates of 30-day complications relative to younger patients. Specifically, patients 65 years and older were significantly more likely to experience readmission, non-home discharge, overall medical complications, pneumonia, urinary tract infections, and myocardial infarction (Table 2).
Table 2Short-Term Outcome Complication Measures and Complication Rates Prior to Matching
| 40-65 Years Old [n = 12,404] | 65+ Years Old [n = 5,476] | P | |||
|---|---|---|---|---|---|
| Readmission | 115 | 0.9% | 84 | 1.5% | <.001 |
| Reoperation | 35 | 0.3% | 12 | 0.2% | .448 |
| Non-home discharge | 76 | 0.6% | 51 | 0.9% | .019 |
| Mortality | 4 | 0.03% | 1 | 0.02% | .606 |
| Surgical complications | |||||
| Overall | 14 | 0.1% | 4 | 0.07% | .610 |
| Superficial surgical site infection | 8 | 0.06% | 2 | 0.04% | .733 |
| Deep surgical site infection | 4 | 0.03% | 1 | 0.02% | >.999 |
| Dehiscence | 2 | 0.02% | 0 | 0.0% | >.999 |
| Bleeding | 1 | 0.01% | 2 | 0.04% | .224 |
| Medical complications | |||||
| Overall | 79 | 0.6% | 70 | 1.3% | <.001 |
| Wound infection | 5 | 0.04% | 2 | 0.04% | >.999 |
| Pneumonia | 12 | 0.1% | 16 | 0.3% | .002 |
| Reintubation | 4 | 0.03% | 4 | 0.07% | .259 |
| Failure to wean intubation | 3 | 0.02% | 3 | 0.05% | .379 |
| Pulmonary embolism | 29 | 0.2% | 11 | 0.2% | .760 |
| Renal insufficiency | 2 | 0.02% | 2 | 0.04% | .591 |
| Renal failure | 1 | 0.01% | 2 | 0.04% | .224 |
| Urinary tract infection | 15 | 0.1% | 19 | 0.3% | .001 |
| Cerebrovascular accident | 2 | 0.02% | 2 | 0.04% | .591 |
| Cardiac arrest | 3 | 0.02% | 1 | 0.02% | >.999 |
| Myocardial infarction | 7 | 0.06% | 12 | 0.2% | .002 |
| Deep venous thromboembolism | 15 | 0.1% | 8 | 0.1% | .665 |
| Systemic sepsis | 4 | 0.03% | 4 | 0.07% | .259 |
| Septic shock | 1 | 0.01% | 2 | 0.04% | .224 |
Exact matching yielded a subgroup of 9,210 patients (e.g., 4,605 patients in each age group). After matching, patients from both age groups had identical distributions of various potential predictor variables—sex, body mass index (BMI), American Society of Anesthesiologists (ASA) classification, and medical comorbidities—diabetes, smoking, chronic obstructive pulmonary disease (COPD), ascites, congestive heart failure (CHF), hypertension, renal failure, dialysis, chronic steroid use, bleeding disorder, and transfusion within 48 hours of surgery (Table 3).
Table 3Demographics, Patient Variables, and Comorbidity Burden of the Cohort After Matching
| 40-65 Years Old [n = 4,605] | 65+ Years Old [n = 4,605] | P | |||
|---|---|---|---|---|---|
| Sex | |||||
| Male | 2,575 | (55.9%) | 2,575 | (55.9%) | >.999 |
| Female | 2030 | (44.1%) | 2030 | (44.1%) | >.999 |
| BMI (kg/m2) | |||||
| Underweight | 26 | (0.6%) | 26 | (0.6%) | >.999 |
| Normal | 750 | (16.3%) | 750 | (16.3%) | >.999 |
| Overweight | 1,779 | (38.6%) | 1,779 | (38.6%) | >.999 |
| Obese Class I | 1,236 | (26.8%) | 1,236 | (26.8%) | >.999 |
| Obese Class II | 541 | (11.7%) | 541 | (11.7%) | >.999 |
| Obese Class III | 274 | (6.0%) | 274 | (6.0%) | >.999 |
| Comorbidities | |||||
| Diabetes | 871 | (18.9%) | 871 | (18.9%) | >.999 |
| Smoking | 352 | (7.6%) | 352 | (7.6%) | >.999 |
| COPD | 92 | (2.0%) | 92 | (2.0%) | >.999 |
| Ascites | 0 | (0.0%) | 0 | (0.0%) | >.999 |
| Congestive heart failure | 0 | (0.0%) | 0 | (0.0%) | >.999 |
| Hypertension | 2,685 | (58.3%) | 2,685 | (58.3%) | >.999 |
| Renal failure | 0 | (0.0%) | 0 | (0.0%) | >.999 |
| Dialysis | 0 | (0.0%) | 0 | (0.0%) | >.999 |
| Chronic steroid use | 62 | (1.3%) | 62 | (1.3%) | >.999 |
| Bleeding disorder | 31 | (0.7%) | 31 | (0.7%) | >.999 |
| Transfusion within 48 hours | 0 | (0.0%) | 0 | (0.0%) | >.999 |
| ASA Class | |||||
| Class 1 (No disturbance) | 156 | (3.4%) | 156 | (3.4%) | >.999 |
| Class 2 (Mild disturbance) | 2,607 | (56.6%) | 2,607 | (56.6%) | >.999 |
| Class 3 (Severe disturbance) | 1,827 | (39.7%) | 1,827 | (39.7%) | >.999 |
| Class 4+ (Life threatening) | 15 | (0.3%) | 15 | (0.3%) | >.999 |
| Operative time | |||||
| Under 60 minutes | 1,312 | (28.5%) | 1,312 | (28.5%) | >.999 |
| Between 60-89 minutes | 1,585 | (34.4%) | 1,585 | (34.4%) | >.999 |
| 90+ minutes | 1,708 | (37.1%) | 1,708 | (37.1%) | >.999 |
ASA, American Society of Anesthesiologists; BMI, body mass index; COPD, chronic obstructive pulmonary disease.
Older patients were more likely relative to the matched younger cohort to experience 30-day unplanned readmission (P = .035) and overall medical complications (P = .036), with pneumonia (P = .089) having the lowest associated P value of any specific medical complication. However, patients in this older cohort were not more likely to have increased rates of most 30-day complication measures, including mortality (P = .250), reoperation (P = .449), non-home discharge (P = .255), overall surgical complications (P = .146), and several medical complications, including myocardial infarction (P = .165), deep venous thromboembolism (P = .206), pulmonary embolism (P = .196), and urinary tract infection (P = .220), among others (Table 4).
Table 4Short-Term Outcome Complication Measures and Complication Rates After Matching
| 40-65 Years Old [n = 4,605] | 65+ Years Old [n = 4,605] | P | |||
|---|---|---|---|---|---|
| Readmission | 44 | (1.0%) | 66 | (1.4%) | .035 |
| Reoperation | 16 | (0.3%) | 12 | (0.3%) | .449 |
| Non-home discharge | 27 | (0.6%) | 36 | (0.8%) | .255 |
| Mortality | 3 | (0.1%) | 0 | (0.0%) | .250 |
| Surgical Complications | |||||
| Overall | 8 | (0.2%) | 3 | (0.1%) | .146 |
| Superficial surgical site infection | 5 | (0.1%) | 1 | (0.0%) | .219 |
| Deep surgical site infection | 1 | (<0.1%) | 1 | (0.0%) | >.999 |
| Dehiscence | 1 | (<0.1%) | 0 | (0.0%) | >.999 |
| Bleeding | 1 | (<0.1%) | 1 | (0.0%) | >.999 |
| Medical Complications | |||||
| Overall | 32 | (0.7%) | 51 | (1.1%) | .036 |
| Wound infection | 3 | (0.1%) | 2 | (0.0%) | .655 |
| Pneumonia | 5 | (0.1%) | 12 | (0.3%) | .089 |
| Reintubation | 0 | (0.0%) | 4 | (0.1%) | .125 |
| Failure to wean intubation | 0 | (0.0%) | 3 | (0.1%) | .250 |
| Pulmonary embolism | 10 | (0.2%) | 5 | (0.1%) | .196 |
| Renal insufficiency | 0 | (0.0%) | 1 | (0.0%) | >.999 |
| Renal failure | 1 | (0.0%) | 2 | (0.0%) | >.999 |
| Urinary tract infection | 9 | (0.2%) | 15 | (0.3%) | .220 |
| Cerebrovascular accident | 1 | (0.0%) | 1 | (0.0%) | >.999 |
| Cardiac arrest | 2 | (0.0%) | 1 | (0.0%) | .564 |
| Myocardial infarction | 4 | (0.1%) | 9 | (0.2%) | .165 |
| Deep venous thromboembolism | 3 | (0.1%) | 7 | (0.2%) | .206 |
| Systemic sepsis | 1 | (0.0%) | 3 | (0.1%) | .317 |
| Septic shock | 0 | (0.0%) | 2 | (0.0%) | .157 |
Upon performing a binary logistic regression model, likelihood ratios for various 30-day complication measures after matching were also calculated as a function of predictor variables including age (5-year unit increases in age), ASA classification, BMI/obesity classification, and sex. After matching, readmission was 1.154 [1.036–1.285] times more likely for each 5-year increase in age. A preoperative hypertension diagnosis or ASA 3+ classification resulted in 2.097 (1.314–3.346) and 1.963 (1.321–2.919) times higher likelihood of short-term readmission. Reoperation and non-home discharge rates were 4.012 (1.752–9.185) and 2.806 (1.407–5.597) times more likely if patients were classified as ASA 3+ or Obesity Class III, respectively. Overall medical complications were 1.026 (1.001–1.053), 0.554 (0.356–0.861), 0.214 (0.076–0.603), and 8.454 (1.907–37.484) times more likely for each 5-year increase in age, males, normal weight patients, and ASA 4 patients, respectively (Table 5).
Table 5Risk Factors Associated With Various 30-Day Complication Measures After Matching
| Complication Measure | Likelihood Ratio [95% CI] |
|---|---|
| Readmission | |
| Age (per 5-year unit increase) | 1.154 [1.036–1.285] |
| Hypertension | 2.097 [1.314–3.346] |
| ASA 3+ | 1.963 [1.321–2.919] |
| Reoperation | |
| ASA 3+ | 4.012 [1.752–9.185] |
| Non-home discharge | |
| Obesity Class III | 2.806 [1.407–5.597] |
| Mortality | N.S. |
| Surgical complications | N.S. |
| Medical complications | |
| Age (per 5-year unit increase) | 1.026 [1.001–1.053] |
| Male | 0.554 [0.356–0.861] |
| Normal weight | 0.214 [0.076–0.603] |
| ASA 4 | 8.454 [1.907–37.484] |
ASA, American Society of Anesthesiologists; CI, confidence interval; N.S., not significant.
Discussion
In this matched cohort study, patients 65 years and older experienced a higher rate of 30-day unplanned readmission and overall medical complications following elective arthroscopic rotator cuff repair. Age as a continuous variable was found to be predictive of 30-day unplanned readmission and overall medical complications. However, these older patients did not have significantly worse rates of other 30-day complication measures, including mortality, reoperation, return to the OR, and non-home discharge—as well as a variety of specific medical/surgical complications. After controlling for several potential confounding variables in a logistic regression model, high ASA classification was found to be a particularly strong predictor of various complication measures, including readmission, reoperation, and medical complications. Overall, the results of this study support our hypothesis that patients aged 65 years and older will have similar short-term complication rates following arthroscopic rotator cuff repair relative to patients between 40-65 years-old. Our findings stand despite patients 65 years and older being more likely to present for surgery as ASA Class 3 and 4+, as well as with various comorbidities. Findings of significant differences in readmission and overall medical complication rates between the two cohorts after matching are of uncertain clinical significance and may have stemmed from more sensitive readmission measures taken due to a patient’s higher age, as well as increased rates of complications unrelated to the arthroscopic rotator cuff repair procedure.
This study found significant differences in both unmatched preoperative patient variables and comorbidities, as well as matched postoperative complications. Older patients (65+ years old) were more likely to present with poorer overall health profiles, as characterized by their higher proportion of ASA Class 3 and 4+ categorization and comorbidities, including diabetes, COPD, congestive heart failure, hypertension, chronic steroid use, and bleeding disorder. While younger patients (40-65 years old) presented with lower, healthier ASA classifications, they were more likely to present with Class II and Class III obesity and were also more likely to be smokers.
Previous studies
10
,17
have provided contrasting conclusions regarding the incidence and significance of short-term complications following rotator cuff repair in elderly patients. Padaki et al. found that patients over 65 had nearly double the odds of having a postoperative complication following arthroscopic rotator cuff repair, with three times the risk of respiratory complications six times the risk of a urinary tract infection (UTI).10
These findings align quite closely with those of our study prior to matching, as our study found double the odds of overall postoperative medical complications and three times the odds of both pneumonia and UTI incidence. In contrast, Witney-Lagen et al. found that postoperative complication rates did not differ between patient cohorts older or younger than 75.17
Our study found that while patients over 65 years old were at increased risk for short-term complications following arthroscopic rotator cuff repair relative to younger patients, the risk of age as an independent risk factor may be overstated per our findings after matching. After matching, the only differences in short-term complications were that older patients had a slightly higher incidence of unplanned readmission and overall medical complications within 30 days of surgery. Other studies have identified advanced age as an important risk factor for unplanned hospital readmission18
, 19
, 20
, 21
and medical complications, including pneumonia22
, 23
, 24
following a variety of orthopaedic procedures, with COPD, male sex, and dependent functional status also emerging as top predictors of postoperative complications.19
,22
, 23
, 24
Per our study, complications for arthroscopic rotator cuff repair in patients 65 years and older may be most contingent on other predictor variables, including ASA and BMI classification rather than these COPD, sex, functional status, or age alone, underscoring the need for detailed patient selection protocols.The decision to pursue surgery for rotator cuff pathology is multifactorial and should consider both short- and long-term outcomes and complication rates. This decision is further complicated by advanced age. The risk factors of decreased vascularization and limited neovascularization potential in elderly patients and their effect on healing potential
25
are of marked importance in this decision. The risk of retear, which occurs in ∼20-40% of patients after surgical repair,26
,27
should also be considered, especially given the association between increased patient age and recurrent tear.26
Additionally, there are important quality-of-life considerations to take into account; these not only include pain and functional disability, but also psychological complications, including depression, anxiety, and sleeping disorders.7
As such, there is still some disagreement regarding elderly patients undergoing elective arthroscopic rotator cuff repair. Short-term complications, the focus of this paper, and debate over their significance are factors that require physician and patient discussion during the consultation period.Appropriate consideration during the patient selection process should also be given to rotator cuff tear characteristics, including tear size, type, fatty infiltration, and planned repair constructs. For instance, Ok et al. divided massive rotator cuff tears into three types: anterosuperior, posterosuperior, and anteroposterior—and found distinctive postoperative outcomes, including retear rate relative to each tear category.
28
The positive relationship between the amount of fatty infiltration in rotator cuff muscles and postoperative complications, including nonhealing and retear has also been studied and affirmed repeatedly,29
, 30
, 31
, 32
, 33
with preoperative grading of fatty infiltration now common practice using the Goutallier classification system.29
As opposed to arthroplasties that replace degenerative tissue with implants, tendon-to-bone healing in rotator cuff repair does not have this option, placing considerable emphasis on the importance of repair constructs used during the procedure. Although this study did not differentiate between different types of repair constructs, current literature supports the notion that double-row repair constructs optimize tendon-to-bone healing and restoration of shoulder strength during rotator cuff repair.34
Other studies have begun to assess the safety of elective orthopaedic procedures in the elderly population specifically. It has been shown that total hip arthroplasty remains a safe and viable treatment modality in patients over 90 years old (nonagenarians).
35
Another study analyzed risks specific to nonagenarians undergoing elective total joint arthroplasty and recommended that while comorbidities did not differ significantly between nonagenarians and younger patients, complications, including longer mean length-of-stay and higher readmission rate, were more characteristic of this cohort and needed to be addressed in a shared preoperative decision-making process.36
Further studies on elderly patients undergoing primary knee arthroplasty and lumbar spine surgery found similar trends in these complications and highlighted the role of the low BMI and decreased functional status as risk factors.37
,38
Collectively, these studies underscore the need for thoughtful preoperative discussions and further research on the critical balance between short-term safety and long-term quality-of-life.Future research should include long-term prospective studies, as well as studies reporting patient-reported outcomes to effectively capture postoperative pain and functional outcomes. Studies that include patient-reported outcomes should also aim to provide insight into potential discrepancies in the statistical versus clinical significance of the significant variables highlighted in this study. Future research should also continue to probe the relative strength of various predictor variables, including age, ASA, and BMI/obesity classification, sex, and various comorbidities in order to inform proper patient selection methods.
Limitations
Limitations of this study include its retrospective nature and limited data capture, which only included 30-day postoperative data. Furthermore, variables specific to the rotator cuff pathology, such as acuity or amount of fatty atrophy, as well as specifics of the anesthesia and surgical intervention, are not available through the database. Additional variables, including tear size, type, and repair constructs, were not included in this study.
Conclusions
In this matched cohort study, patients 65 years and older experienced a higher rate of 30-day unplanned readmission and overall medical complications following elective arthroscopic rotator cuff repair relative to patients under 65. However, these older patients did not have significantly worse rates of other 30-day complication measures, including mortality, reoperation, return to the OR, and non-home discharge.
Supplementary Data
- ICMJE author disclosure forms
References
- The next four decades: The older population in the United States: 2010 to 2050.US Department of Commerce, Economics and Statistics Administration, Washington, DC2010
- Osteoarthritis.Lancet. 2019; 393: 1745-1759
- Rotator cuff disease: Treatment options and considerations.Sports Med Arthrosc Rev. 2018; 26: 129-133
- Early versus delayed mobilization following rotator cuff repair.J Clin Orthop Trauma. 2019; 10 (Erratum in: J Clin Orthop Trauma 2020;11:1176): 257-260
- Degenerative rotator cuff tear, repair or not repair? A review of current evidence.Ann R Coll Surg Engl. 2020; 102: 248-255
- The relationship between functional disability and health-related quality of life in patients with a rotator cuff tear.Disabil Rehabil. 2012; 34: 2071-2075
- Is shoulder pain for three months or longer correlated with depression, anxiety, and sleep disturbance?.J Shoulder Elbow Surg. 2013; 22: 222-228
- Arthroscopic repair of massive rotator cuff tears with stage 3 and 4 fatty degeneration.Arthroscopy: The Journal of Arthroscopic & Related Surgery. 2007; 23: 347-354
- Rotator cuff surgery in patients older than 75 years with large and massive tears.Journal of shoulder and elbow surgery. 2017; 26: 265-272
- The effect of age on short-term postoperative complications following arthroscopic rotator cuff repair.JSES Open Access. 2019; 3: 194-198
- Development and evaluation of the universal ACS NSQIP surgical risk calculator: A decision aid and informed consent tool for patients and surgeons.J Am Coll Surg. 2013; 217: 833-842.e833
- Toward robust information: Data quality and inter-rater reliability in the American College of Surgeons National Surgical Quality Improvement Program.J Am Coll Surg. 2010; 210: 6-16
- (Published 2020. Accessed January 23, 2021)
- A propensity score-matched analysis comparing outpatient and short-stay hospitalization to standard inpatient hospitalization following total ankle arthroplasty.J Orthop Surg Res. 2020; 15: 1-9
- Differences in 30-day outcomes between inpatient and outpatient total elbow arthroplasty (TEA).J Shoulder Elbow Surg. 2020; 29: 2640-2645
- Single-level posterolateral fusion (PLF) alone and posterior interbody fusion (PLIF/TLIF) alone lead to a decreased risk of short-term complications compared to combined PLF with PLIF/TLIF procedures: A matched analysis.Spine. 2020; 45: E1391-E1399
- Do elderly patients gain as much benefit from arthroscopic rotator cuff repair as their younger peers?.J Shoulder Elbow Surg. 2019; 28: 1056-1065
- Incidence of and risk factors for 30-day readmission following elective primary total joint arthroplasty: Analysis from the ACS-NSQIP.J Arthroplasty. 2013; 28: 1499-1504
- Causes and risk factors for 30-day unplanned readmissions after lumbar spine surgery.Spine. 2014; 39: 761-768
- Impact of age on 30-day complications after adult deformity surgery.Spine. 2018; 43: 120-126
- Reasons for readmission following primary total shoulder arthroplasty.Am J Orthop (Belle Mead, NJ). 2018; 47
- Incidence, risk factors, and clinical implications of pneumonia after surgery for geriatric hip fracture.J Arthroplasty. 2018; 33: 1552-1556.e1551
- Incidence and risk factors for pneumonia following anterior cervical decompression and fusion procedures: An ACS-NSQIP study.Spine J. 2016; 16: 335-342
- Incidence, risk factors, and clinical implications of pneumonia following total hip and knee arthroplasty.J Arthroplasty. 2017; 32: 1991-1995.e1991
- Age-related structural alterations of skeletal muscles and associated capillaries.Angiogenesis. 2020; 23: 79-82
- Arthroscopic rotator cuff repair: Prospective functional outcome and repair integrity at minimum 2-year follow-up.J Shoulder Elbow Surg. 2007; 16: 579-585
- Rotator cuff repair in patients over 75 years of age: Clinical outcome and repair integrity.Clinics Orthop Surg. 2016; 8: 420
- Clinical relevance of classifying massive rotator cuff tears: Results based on functional and radiological findings after arthroscopic repair.Am J Sports Med. 2017; 45: 157-166
- 2003. Influence of cuff muscle fatty degeneration on anatomic and functional outcomes after simple suture of full-thickness tears.J Shoulder Elbow Surg. 2003; 12: 550-554
- Fatty infiltration and atrophy of the rotator cuff do not improve after rotator cuff repair and correlate with poor functional outcome.Am J Sports Med. 2007; 35: 719-728
- Arthroscopic rotator cuff repair using a suture bridge technique: Is the repair integrity actually maintained?.Am J Sports Med. 2011; 39: 2108-2116
- Rotator cuff re-tear or non-healing: Histopathological aspects and predictive factors.Knee Surg Sports Traumatol Arthrosc. 2011; 19: 1588-1596
- Relationship between muscle fatty infiltration and the biological characteristics and stimulation potential of tenocytes from rotator cuff tears.J Orthop Res. 2014; 32: 129-137
- Arthroscopic rotator cuff repair in 2019: Linked, double row repair for achieving higher healing rates and optimal clinical outcomes.Arthroscopy. 2019; 35: 2749-2755
- Outcomes of elective total hip arthroplasty in nonagenarians and centenarians.J Arthroplasty. 2020; 35: 2149-2154
- Total joint arthroplasty in nonagenarians: What are the risks?.J Arthroplasty. 2015; 30: 2102-2105.e2101
- The impact of older age on patient outcomes following primary total knee arthroplasty.Bone Joint J. 2018; 100: 1463-1470
- Complications and readmission after lumbar spine surgery in elderly patients: an analysis of 2,320 patients.Spine J. 2017; 17: 1106-1112
Article info
Publication history
Published online: May 24, 2022
Accepted:
April 21,
2022
Received:
February 17,
2021
Footnotes
The authors report the following potential conflicts of interest or sources of funding: V.K.T. reports personal and consulting fees from Smith & Nephew, outside the submitted work. Full ICMJE author disclosure forms are available for this article online, as supplementary material.
Identification
Copyright
© 2022 The Authors. Published by Elsevier Inc. on behalf of the Arthroscopy Association of North America
User license
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0) | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution – NonCommercial – NoDerivs (CC BY-NC-ND 4.0)
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
- Text & data mine
- Translate the article (private use only, not for distribution)
- Reuse portions or extracts from the article in other works
Not Permitted
- Sell or re-use for commercial purposes
- Distribute translations or adaptations of the article
Elsevier's open access license policy
