Abstract
Background
Understanding the impact of situational variables on surgical recovery can improve outcomes in total hip arthroplasty (THA). Literature examining hospital outcomes by season remains inconclusive, with limited focus on patient experience. The aim of this study is to investigate if there are differences in hospital and patient-reported outcomes measures (PROMS) after THA depending on the season of the index procedure to improve surgeon preoperative counseling.
Methods
A retrospective chart review was performed on patients undergoing primary THA at a single large academic center between January 2013 and August 2020. Demographic, operative, hospital, and PROMs were gathered from the institutional electronic medical record and our institutional joint replacement outcomes database.
Results
6418 patients underwent primary THA and met inclusion criteria. Of this patient population, 1636 underwent surgery in winter, 1543 in spring, 1811 in summer, and 1428 in fall. PROMs were equivalent across seasons at nearly time points. The average age of patients was 65 (+/- 10) years, with an average BMI of 29.3 (+/- 6). Rates of complications including ED visits within 30 days, readmission within 90 days, unplanned readmission, dislocation, fracture, or wound infection were not significantly different by season (P > .05).
Conclusion
Our findings indicate no differences in complications and PROMs at 1 year in patients undergoing THA during 4 distinct seasons. Notably, patients had functional differences at the second follow-up visit, suggesting variation in short-term recovery. Patients could be counseled that they have similar rates of complications and postoperative recovery regardless of season.
Keywords: arthroplasty, seasonal variation, patient-reported outcome measures, patient-reported outcomes measures, total hip arthroplasty
Introduction
A plethora of modifiable and nonmodifiable preoperative factors have been examined to optimize outcomes for patients undergoing total joint arthroplasty (TJA). 1 However, little attention has been paid to the role of situational variables in preoperative decision making prior to joint replacement. There is a growing body of orthopedic literature regarding the effect of season on postoperative outcomes. Several of these studies examine the “July effect” of new fellows and residents, with no significant differences reported.2-4 However, complications following total knee arthroplasty (TKA) have been found to be higher during the summer months 5 with increased risk for infections found during warmer weather.6-8 In a large database study, total hip arthroplasty (THA) and TKA procedures performed in December were shown to have lower rates of complications with decreased lengths of stay postoperatively. 9 Alternatively, another study noted increased rates of periprosthetic infection following THA occur during the winter, 10 although this was not demonstrated in a population of unicompartmental knee arthroplasty patients. 11 No effect of season has been noted on post-operative wound complications after trauma to the foot, ankle or lower leg. 12 Another study found no difference in seasonal variation of complications 30 days after THA. 13 Taken as a whole, the body TJA literature remains inconclusive on the role of season of index surgery on outcomes. The majority of work within this field has focused on whether season of surgery impacts complications; there to the best of our knowledge this is the first study to investigate whether season of surgery effects patient-reported outcome measures (PROMs) and satisfaction following TJA.
Preoperative and post-operative PROMs are a standard practice in TJA. 14 These measures evaluate patients’ perceptions of their health and experiences, and are essential to providing patient-centered care. 15 Additionally, as measures of patient satisfaction, PROMs are increasingly utilized as indicators of health care quality, with the ability to influence health care policy and decrease inequalities within care systems. 16 Understanding the factors that impact PROMs are important for maximizing patient satisfaction and identifying areas for quality improvement. We hypothesize that season of index surgery may impact PROMs following THA. This effect may be impacted by variation in complication rate by season, if seen, with higher complication rates associated with lower PROMs. Yet additional environmental and behavioral may also play a role. Winter months in northeastern regions of the United States (where our institution is located) encourage a sedentary lifestyle, which may adversely impact patient postoperative mobility and recovery.17,18 Furthermore, increased rates of depression and Seasonal Affective Disorder occur during winter months19,20; such psychological conditions increase complications and decrease satisfaction among TJA patients, 21 and may similarly limit postoperative mobility and therefore recovery. Conversely, the summer months in northeastern United States are notable for increased ambient temperatures, a factor which increases microbe survivorship, but which also allows for increased patient post-operative activity and ability to be outside. 22 Thus, the role of season and its impact on patient satisfaction in recovery is poorly understood, with an unclear expected relationship between season of surgery and PROMs following THA.
The aim of this study is to investigate if there are differences in hospital and patient-reported outcomes after THA depending on the season of the index procedure performed in the northeastern United States. Better understanding of how season may affect patient-reported outcomes following THA will allow improved surgeon preoperative counseling prior to undergoing arthroplasty as an elective procedure.
Material and Methods
Patient Selection and Data Collection
A retrospective chart review was performed on patients undergoing primary total hip arthroplasty at a single large academic center in northern New England with 4 distinct seasons between January 1, 2013 and August 31, 2020. Institutional review board approval (IRB NET ID 1981001-1) was obtained, and the project was deemed not to constitute human subjects’ research. Inclusion criteria was set for patients over the age of 18 who received an anterior-based muscle-sparing THA, 23 the approach commonly used at our hospital. Patients undergoing revision THA, patients who had a preoperative diagnosis of septic arthritis, and patients with a discharge disposition of psychiatric hospital or expired were excluded. All cases performed by the only 3 arthroplasty surgeons performing THA were included for analysis. The primary outcomes of interest were the demographics, surgical, and dispositional outcomes of patients undergoing THA in different calendar seasons. Seasons were defined using the standard meteorological definition of 4 3 month periods, winter (December 1—February 28), spring (March 1—May 31), summer (June 1—August 31), and fall (September 1—November 30). 24
Demographic, operative, and hospital outcome data were obtained from the institutional electronic medical record. Demographic information collected included sex, age, American Society of Anesthesiology (ASA) score, preoperative diagnosis (degenerative joint disease [DJD]/osteoarthritis [OA], avascular necrosis [AVN], and fracture), and public or private insurance type. Operative variables collected included arthroplasty fixation (press fit vs cemented), procedure type (primary vs conversion), anesthesia time in minutes, procedure duration in minutes, estimated blood loss (EBL) in milliliters, transfusion requirement, and presence of an intra-operative complication. Hospital-reported outcomes included discharge disposition (to home, home health care service, skilled nursing facility, or rehabilitation facility), emergency department (ED) visit (within 30 days), average days after surgery for ED visit, readmission (within 90 days), average days after surgery for readmission, planned or unplanned readmission, presence of any complication (as well as specifically surgical complications), myocardial infarction (within 7 days), pneumonia (within 7 days), surgical site complication (within 30 days), pulmonary embolism (within 30 days), death (within 30 days), fracture (within 90 days), dislocation (within 90 days), mechanical complication (within 90 days), and joint or wound infection (within 90 days). Initially, 6421 patients were included. Patient-reported outcomes collected included preoperative, 6 weeks, 6 months, and 1 year Single Assessment Numeric Evaluation (SANE), pain, Hip Disability and Osteoarthritis Outcome Score Joint Replacement (HOOS JR), University of California, Los Angeles (UCLA), and Patient- Reported Outcomes Measurement Information System (PROMIS) mental/physical and satisfaction scores, all collected from an institutional joint replacement outcomes database. After excluding patients who were undergoing revision THA, patients who had a pre-operative diagnosis of septic arthritis, and patients with a discharge disposition of expired or psychiatric hospital, the data included 6418 patients.
Data Analysis
To assess the relationship between seasonality and our outcomes of interest, all demographic, patient-reported, and hospital-reported variables were analyzed with respect to season. Pearson’s Chi-square test was used for all normally distributed categorical variables, and the Wilcoxon rank sum test or Fisher’s exact test were used for non-normally distributed categorical variables and all continuous variables. Q-values were reported in all outcomes tables to correctly adjust for the False Discovery Rate for multiple testing.
Univariate regression models were created to analyze the relationship between patient- and hospital-reported outcomes and all covariates (cemented status, both hips, sex, age, BMI category, insurance category, and anesthesia type). Linear and logistic regression approaches were used according to the respective outcome (linear for continuous outcomes, logistic for binary outcomes). Final linear and logistic multivariable regression models were created using a combination of clinical expertise and purposeful selection (P < .2) with respect to each individual outcome; adjustment was made for year of index surgery. Any patients with missing data were excluded from regression analyses.
Modeling of patient-reported outcomes measures (PROMs) was limited by missing data (>50% of patients were missing 1 or more outcome values) impacting the accuracy and generalizability of PROMs regression model results. Descriptive analysis showing differences across season was utilized to assess PROMs that were significantly different on initial analysis. All analysis was done on R version 4.2.1.
Results
From January 1, 2013 to August 31, 2020, 6418 patients underwent primary THA and met inclusion criteria. Of this patient population, 1636 underwent surgery in winter, 1543 in spring, 1811 in summer, and 1428 in fall, respectively. The average age of patients was 65 (+/- 10) years, with an average BMI of 29.3 (+/- 6). Women comprised 55% of the patient population with 96% of patients undergoing uncemented THA. Most surgeries performed were primary procedures (99%), with slightly more unilateral right THA (53%) (Table 1). Between the seasonal groups, demographic characteristics remained largely the same, without significant difference in age, BMI, or insurance category. Sex varied slightly (55% female in the winter group, 54% in spring, 58% in summer, 53% in fall). As did cementation (between 3.1% in fall and 5.2% in spring).
Table 1.
Demographic Information.
| Characteristic | Overall, N = 6,418 a | Winter, N = 1,636 a | Spring, N = 1,543 a | Summer, N = 1,811 a | Fall, N = 1,428 a |
|---|---|---|---|---|---|
| Cemented (yes) | 246 (3.8%) | 59 (3.6%) | 81 (5.2%) | 62 (3.4%) | 44 (3.1%) |
| Primary procedure | 6348 (99%) | 1611 (98%) | 1533 (99%) | 1793 (99%) | 1411 (99%) |
| Laterality (right) | 3426 (53%) | 878 (54%) | 799 (52%) | 993 (55%) | 756 (53%) |
| Both hips replaced | 1670 (26%) | 419 (26%) | 381 (25%) | 466 (26%) | 404 (28%) |
| Sex (female) | 3532 (55%) | 903 (55%) | 831 (54%) | 1042 (58%) | 756 (53%) |
| Age at discharge | 65 (10) | 65 (10) | 65 (10) | 66 (10) | 65 (10) |
| Weight (kg) | 85 (20) | 85 (20) | 85 (21) | 85 (20) | 84 (20) |
| Height (m) | 1.70 (.10) | 1.70 (.10) | 1.70 (.10) | 1.69 (.10) | 1.70 (.10) |
| BMI | 29.3 (6.1) | 29.3 (5.7) | 29.4 (6.2) | 29.4 (6.4) | 29.0 (5.8) |
| BMI category | |||||
| Underweight | 64 (1.0%) | 12 (.7%) | 22 (1.4%) | 19 (1.0%) | 11 (.8%) |
| Healthy weight | 1531 (24%) | 375 (23%) | 345 (22%) | 450 (25%) | 361 (25%) |
| Overweight | 2268 (35%) | 588 (36%) | 560 (36%) | 604 (33%) | 516 (36%) |
| Obese | 2555 (40%) | 661 (40%) | 616 (40%) | 738 (41%) | 540 (38%) |
| Insurer category | |||||
| Government | 3559 (56%) | 870 (54%) | 852 (56%) | 1013 (56%) | 824 (58%) |
| Private | 2796 (44%) | 756 (46%) | 669 (44%) | 784 (44%) | 587 (42%) |
| Year of surgery | |||||
| 2013 | 619 (9.6%) | 147 (9.0%) | 148 (9.6%) | 171 (9.4%) | 153 (11%) |
| 2014 | 765 (12%) | 209 (13%) | 191 (12%) | 166 (9.2%) | 199 (14%) |
| 2015 | 775 (12%) | 190 (12%) | 194 (13%) | 192 (11%) | 199 (14%) |
| 2016 | 899 (14%) | 203 (12%) | 243 (16%) | 253 (14%) | 200 (14%) |
| 2017 | 927 (14%) | 260 (16%) | 219 (14%) | 244 (13%) | 204 (14%) |
| 2018 | 969 (15%) | 238 (15%) | 258 (17%) | 254 (14%) | 219 (15%) |
| 2019 | 946 (15%) | 224 (14%) | 236 (15%) | 232 (13%) | 254 (18%) |
| 2020 | 518 (8.1%) | 165 (10%) | 54 (3.5%) | 299 (17%) | 0 (0%) |
an (%); Mean (SD).
The surgical variables anesthesia type, anesthesia time, procedure duration, and room duration showed some variation between the seasons. ASA rating, blood transfusion, EBL and intra-operative complication rates did not differ among groups (Table 2), nor did the distribution of preoperative diagnoses and post-operative dispositions. There was no significant difference in hospital-reported outcomes between seasonal groups (any complication P-value >.05) (Table 3). Among seasonal groups, rates of such complications of ED visit within 30 days, readmission within 90 days, unplanned readmission, myocardial infarction or pneumonia (within 7 days), surgical site complication, pulmonary embolism, or death (within 30 days), as well as dislocation, fracture, mechanical complication, or joint or wound infection (within 90 days) were not significantly different by season. Since seasonality was not crudely associated with any outcomes, outcomes for univariate analyses were selected using highest number of events (ie the most common outcomes across all seasons).
Table 2.
Surgical Variables and Disposition.
| Characteristic | Overall, N = 6,418 a | Winter, N = 1,636 a | Spring, N = 1,543 a | Summer, N = 1,811 a | Fall, N = 1,428 a |
|---|---|---|---|---|---|
| ASA rating | 2 (1) | 2 (0) | 2 (1) | 2 (1) | 2 (1) |
| Unknown | 869 | 242 | 168 | 274 | 185 |
| Pre-op diagnosis | |||||
| DJD/OA | 6182 (97%) | 1574 (97%) | 1489 (97%) | 1740 (97%) | 1379 (97%) |
| AVN | 94 (1.5%) | 24 (1.5%) | 21 (1.4%) | 32 (1.8%) | 17 (1.2%) |
| Fracture | 102 (1.6%) | 29 (1.8%) | 25 (1.6%) | 29 (1.6%) | 19 (1.3%) |
| Unknown | 40 | 9 | 8 | 10 | 13 |
| General anesthesia | 6223 (97%) | 1560 (95%) | 1489 (97%) | 1788 (99%) | 1386 (97%) |
| Unknown | 1 | 0 | 0 | 1 | 0 |
| Anesthesia time (minutes) | 109 (21) | 109 (21) | 110 (22) | 108 (21) | 108 (22) |
| Unknown | 3 | 1 | 1 | 1 | 0 |
| Procedure duration (minutes) | 66 (19) | 67 (19) | 67 (18) | 65 (18) | 65 (19) |
| Room duration (minutes) | 102 (21) | 102 (21) | 103 (21) | 102 (21) | 102 (21) |
| Length of stay (hour) | 34 (20) | 35 (21) | 33 (17) | 34 (22) | 34 (20) |
| Discharge disposition | |||||
| Home or self care | 3842 (60%) | 979 (60%) | 906 (59%) | 1125 (62%) | 832 (58%) |
| Home health care svc | 2100 (33%) | 534 (33%) | 524 (34%) | 558 (31%) | 484 (34%) |
| Skilled nursing facility | 395 (6.2%) | 104 (6.4%) | 96 (6.2%) | 105 (5.8%) | 90 (6.3%) |
| Rehab facility | 81 (1.3%) | 19 (1.2%) | 17 (1.1%) | 23 (1.3%) | 22 (1.5%) |
| EBL (mL) | 221 (78) | 221 (81) | 223 (74) | 221 (80) | 221 (76) |
| Unknown | 1567 | 353 | 345 | 498 | 371 |
| Blood transfusion | 55 (.9%) | 17 (1.0%) | 12 (.8%) | 17 (.9%) | 9 (.6%) |
| Intra-op complication | 19 (.3%) | 4 (.2%) | 3 (.2%) | 7 (.4%) | 5 (.4%) |
an (%); Mean (SD).
Table 3.
Bivariate Analysis of Hospital Outcomes.
| Characteristic | Overall, N = 6,418 a | Winter, N = 1,636 a | Spring, N = 1,543 a | Summer, N = 1,811 a | Fall, N = 1,428 a | P-value b | q-value b |
|---|---|---|---|---|---|---|---|
| ED visit within 30 days | 122 (1.9%) | 35 (2.1%) | 27 (1.7%) | 37 (2.0%) | 23 (1.6%) | .7 | >.9 |
| Average days after surgery for ED visit | 8 (4, 15) | 8 (6, 13) | 7 (4, 14) | 8 (5, 13) | 7 (2, 23) | >.9 | >.9 |
| Readmission within 90 days | 192 (3.0%) | 48 (2.9%) | 37 (2.4%) | 58 (3.2%) | 49 (3.4%) | .4 | >.9 |
| Average days after surgery for readmission | 33 (13, 59) | 22 (10, 57) | 33 (16, 59) | 38 (18, 57) | 41 (16, 67) | .6 | >.9 |
| Surgical or medical complication (surgical) | 56 (67%) | 14 (70%) | 12 (60%) | 16 (70%) | 14 (70%) | .9 | >.9 |
| Readmission unplanned or unplanned prior | 192 (3.0%) | 48 (2.9%) | 37 (2.4%) | 58 (3.2%) | 49 (3.4%) | .4 | >.9 |
| Any complication | 83 (1.3%) | 20 (1.2%) | 20 (1.3%) | 23 (1.3%) | 20 (1.4%) | >.9 | >.9 |
| Myocardial infarction within 7 days | 8 (.1%) | 2 (.1%) | 3 (.2%) | 3 (.2%) | 0 (0%) | .4 | >.9 |
| Pneumonia within 7 days | 3 (<.1%) | 0 (0%) | 1 (<.1%) | 0 (0%) | 2 (.1%) | .10 | >.9 |
| Surgical site complication within 30 days | 4 (<.1%) | 1 (<.1%) | 1 (<.1%) | 0 (0%) | 2 (.1%) | .4 | >.9 |
| Pulmonary embolism within 30 days | 4 (<.1%) | 2 (.1%) | 0 (0%) | 1 (<.1%) | 1 (<.1%) | .6 | >.9 |
| Death within 30 days | 2 (<.1%) | 0 (0%) | 1 (<.1%) | 1 (<.1%) | 0 (0%) | .9 | >.9 |
| Fracture within 90 days | 25 (.4%) | 7 (.4%) | 4 (.3%) | 7 (.4%) | 7 (.5%) | .8 | >.9 |
| Dislocation within 90 days | 10 (.2%) | 3 (.2%) | 4 (.3%) | 3 (.2%) | 0 (0%) | .3 | >.9 |
| Mechanical complication within 90 days | 4 (<.1%) | 2 (.1%) | 1 (<.1%) | 1 (<.1%) | 0 (0%) | .8 | >.9 |
| Joint infection within 90 days | 12 (.2%) | 0 (0%) | 2 (.1%) | 5 (.3%) | 5 (.4%) | .066 | .8 |
| Wound infection within 90 days | 11 (.2%) | 3 (.2%) | 3 (.2%) | 2 (.1%) | 3 (.2%) | .9 | >.9 |
an (%); Mean (SD).
bPearson's Chi-squared test; Kruskal-Wallis rank sum test; Fisher's exact test.
cFalse discovery rate correction for multiple testing.
Univariate analyses were performed evaluating the impact of seasonality on ED visit within 30 days of THA (n = 122) and readmission within 90 days of THA (n = 192). Season was not associated with significant difference in odds ratio of ED visit within 30 days of THA nor readmission within 90 days of THA. Covariates demonstrating P < .2 (age, BMI category, insurance, and anesthesia categories for ED visit model; cemented status, both hips, age, BMI category, insurance, and anesthesia categories for readmission model) were selected for inclusion in multivariable analyses. On multivariable analysis, ED visit within 30 days of THA was not significantly associated with age, BMI category, insurance type, nor anesthesia type. When season was used in the model as a predictor, odds ratios remained insignificant, indicating season is likely not a robust predictor of ED utilization following THA (Table 4). When a multivariable analysis was performed for readmission within 90 days of THA, baseline covariates of both hips replaced, age, BMI category, and insurance type were significant (P = <.05). When season was used in the model as a predictor, only the odds ratios of these covariates remained significant, indicating season is likely not a robust predictor of readmission within 90 days of THA (Table 5).
Table 4.
Multivariate Regression Analyses for ED Visit Within 30 days.
| Characteristic | OR a (95% CI a ) | P-value | q-value b |
|---|---|---|---|
| Season | .7 | .7 | |
| Winter | - | ||
| Spring | .85 (.50, 1.41) | ||
| Summer | 1.01 (.63, 1.63) | ||
| Fall | .77 (.45, 1.32) | ||
| Private insurance | .61 (.41, .88) | .009 | .027 |
| Spinal anesthesia | 2.20 (.97, 4.34) | .058 | .087 |
aOR = Odds Ratio, CI = Confidence Interval.
bFalse discovery rate correction for multiple testing.
Null deviance = 1198; Null df = 6353; Log-likelihood = -593; AIC = 1198; BIC = 1239; Deviance = 1186; Residual df = 6348; No. Obs. = 6354.
Table 5.
Multivariate Regression Analyses for Readmission Within 90 days.
| Characteristic | OR1 (95% CI1) | P-value | q-value a |
|---|---|---|---|
| Season | .3 | .3 | |
| Winter | - | ||
| Spring | .76 (.48, 1.17) | ||
| Summer | 1.04 (.70, 1.54) | ||
| Fall | 1.13 (1.02, 1.06) | ||
| Age at discharge | 1.03 (1.02, 1.06) | <.001 | .002 |
| BMI category | .046 | .061 | |
| Underweight | |||
| Healthy weight | .39 (.16, 1.17) | ||
| Overweight | .33 (.14, .98) | ||
| Obese | .49 (.20, 1.44) | ||
| Private insurance | .51 (.33, .80) | .003 | .006 |
aOR = Odds Ratio, CI = Confidence Interval.
bFalse discovery rate correction for multiple testing.
Null deviance = 1715; Null df = 6353; Log-likelihood = -823; AIC = 1664; BIC = 1725; Deviance 1646; Residual df = 6345; No. Obs. = 6354.
PROMS are reported from preoperative to 1 year postoperatively (Table 4). With this analysis, there were no values identified that demonstrated a statistically significant difference across season, nor met minimal clinical significant difference. Only preoperative UCLA score varied (q < .001). Patient-reported outcome measures of function and satisfaction were equivalent across season of index surgery and time.
Discussion
Demographics and Outcomes
Overall, demographic characteristics, as well as surgical and hospital variables, remained remarkably consistent across season of index surgery. The homogeneity amongst the seasonal groups in most categories supports conclusions that findings by season are not due to otherwise unaccounted-for differences between the groups. There were few statistically significant differences in demographic data observed, including variation in rate of cementation and percentage of female patients by season. Some surgical and hospital variables showed statistically significant differences, including percentage undergoing surgery under general anesthesia, anesthesia time, procedure duration, and room duration, although these are likely not clinically significant. None of the postoperative variables measured were significantly different across season: differences in incidence of complication and readmission within 90 days (among other variables) were not statistically significant. Although it is possible that some patients may have sought care at facilities other than our institution, we feel that early readmission would be directed to the surgeon that performed the original surgery. Additionally, most hospitals within our geographic area are operated under the same health care system and utilize the same electronic medical record, and admissions are referred back to the providing surgeon at our institution, the largest hospital and only Level 1 trauma center in the region.
In contrast to prior work, which implicates both summer and winter as seasons with increased incidence of prosthetic joint infection and complications,25,26 we did not observe differences in incidence of surgical site complication within 30 days nor joint or wound infection within 90 days by season, Due to the rarity of these complications, our study may be underpowered to detect a statistical difference. It is possible that each season confers specific risks and protective factors. Several studies have identified increased temperature as a risk factor for infection following TJA, with hay fever and associated allergies higher during summer months. 27 Winter months in areas experiencing colder ambient air temperatures are associated with increased eczema flares 28 and upper respiratory infections. 29 While environmental factors may increase risk during winter. Aghdassi et al showed that patients undergoing surgery on days with temperature over 20°C had increased risks of surgical site infection compared to those less than 5°C. 30 Furthermore, Fortaleza et al showed an increased risk of 1.3% of SSI for every increase of 1° in average temperature on the day of surgery. 31 Increased temperature during the summer months improves microbe survivorship, which may underly an increased risk for infection. 22 Thus, patients with increased infection risk may consider undergoing THA at a cooler time of year. Conversely, during winter months, patients generally have less sun exposure. This may increase the risk of vitamin D deficiency, which is associated with complications following TJA and occurs at higher prevalence in people undergoing revision arthroplasty. 32 Thus, there is evidence that THA recovery may be impacted in the summer by warmer weather and elevated infection risk, and in the winter by increased in risk for complications. At our institution located in northern New England, with average January temperatures ranging from 13-31 F and average July temperatures ranging from 59-79 F, we did not observe an increased incidence of infection or any complication during the summer or winter months. 33
There has been concern that the increases in infection and complications observed during the summer months are due to the presence of inexperienced trainees—new residents and fellows joining the care team— however this has not been widely supported by the literature.2-4 At our tertiary care institution without an orthopedic training program, patients underwent THA by fellowship-trained attending surgeons without the involvement of trainees. Khanna et al 26 found an increased rate of complications in summer months in teaching and non-teaching hospitals alike. Thus, the trend in our data may support the finding that increased incidence of infections in the summer—when observed— are not driven by trainee involvement but may be the result of a seasonal or temperature effect.
Our data demonstrated no statistical differences in discharge disposition between seasons. In our institution’s northern geographic location, it would be reasonable to consider that it may be more difficult for patients to return home during the winter months, where it can be challenging to perform activities of daily living due to the cold temperatures, ice and snow. However, the finding that patients largely return home (92%–93% on average by season discharged to home/self-care or with home health services) provides evidence for the success of THA, as well as good patient selection of patients who are able to have a successful recovery. Zalikha et al 9 found shorter length of stay and a greater proportion of patients returning home when undergoing arthroplasty in December. They hypothesized that the cultural significance of December holidays may impact discharge disposition planning, and the presence of family support may allow for more rapid return home. Our data showed that patients undergoing THA during summer did have the lowest percentage going to skilled rehab, although by a very small percentage. From this data, patients could be counseled that they have a similar chance to return home postoperatively regardless of season of undergoing THA.
Patient Reported Outcome Measures
Patient reported outcome measures remained mostly equivalent amongst seasonal groups; postoperatively, there were no statistically significant differences in scores of function or satisfaction. Increased rates of depression and Seasonal Affective Disorder may occur during the winter,19,20 which increases complications and decreases satisfaction among TJA patients. 21 A sedentary lifestyle during the winter may adversely impact recovery.17,18 There is evidence that older patients have decreased activity in the winter34,35 which may impact patient recovery to higher levels of activity. Pervasive differences were not observed, but it still may be beneficial to counsel patients that their activity levels may be lower undergoing THA in the winter or spring compared to the summer or fall. With the importance of rehabilitation for TKA, 36 these differences may be increased within that population compared to THA patients. Furthermore, it may be appropriate to have patients who are at higher risk of poor short-term recovery elect to undergo surgery during the warmer months. This data indicates while there may be some differences in recovery, seasonality does not impact patient-reported outcomes measured up to 1 year.
Limitations
There are several notable limitations to this study. With patient-reported outcome surveys, there is potential for incomplete data and bias. Nevertheless, previous studies have found non-responders have similar responses in TJA PROMs. 37 The patients with missing data were excluded from regression analyses, resulting in loss of information due to incompleteness. Additionally, the population encapsulated in our study was racially and ethnically homogenous, limiting generalizability to areas where there is more diversity. While we have categorized based on meteorological seasons, weather patterns may be variable from year to year, altering conditions that patients may face. Reassuringly, average temperatures across years at our institution’s location remained stable with relatively little fluctuation, indicating consistency in the experience of winter, spring, summer, and fall seasons across the years observed; year of surgery was also accounted for in multivariable analysis. 33 Data from only 3 surgeons at 1 center was analyzed in this study, further limiting generalizability.
Conclusion
When evaluating demographics, surgical, and hospital variables, characteristics were similar regardless of season of index surgery. Statistically significant variation in complications were not observed. In evaluation of PROMs, no differences in PROMs measured up to 1 year were found between the 4 seasons in a patient population undergoing THA during 4 distinct seasons. From this data, patients could be counseled that they have similar rates of complications and postoperative recovery regardless of season.
Supplemental Material
Supplemental Material for The Association of Season of Surgery and Patient Reported Outcomes following Total Hip Arthroplasty by Andrew D. Lachance, Catherine Call, Zachary Radford, Henry Stoddard, Callahan Sturgeon, George Babikian, Adam Rana, and Brian J. McGrory in Progress in Physical Geography
Author’s Note: All authors qualify for authorship as per ICMJE guidelines. All authors approve of this submission. We thank the reviewers for the consideration of this manuscript.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material: Supplemental material for this article is available online.
ORCID iD
Andrew D. Lachance https://orcid.org/0000-0003-2666-8327
References
- 1.Odum SM, Johnson NR, Otero JE. Failure to medically optimize before total hip arthroplasty: Which modifiable risk factor is the most dangerous? Arthroplast Today. 2021;10:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bohl DD, Fu MC, Golinvaux NS, Basques BA, Gruskay JA, Grauer JN. The “July effect” in primary total hip and knee arthroplasty: Analysis of 21,434 cases from the ACS-NSQIP database. J Arthroplasty. 2014;29(7):1332. [DOI] [PubMed] [Google Scholar]
- 3.Rockov ZA, Etzioni DA, Schwartz AJ. The July effect for total joint arthroplasty procedures. Orthopedics. 2020;43(6):e543. [DOI] [PubMed] [Google Scholar]
- 4.Tobert DG, Menendez ME, Ring DC, Chen NC. The “July effect” on shoulder arthroplasty: Are complication rates higher at the beginning of the academic year? Archives of Bone and Joint Surgery. 2018;6(4):277. [PMC free article] [PubMed] [Google Scholar]
- 5.Malik AT, Azmat SK, Ali A, Mufarrih SH, Noordin S. Seasonal influence on postoperative complications after total knee arthroplasty. Knee Surgery & Related Research. 2018;30(1):42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Sahtoe APH, Duraku LS, van der Oest MJW, et al. Warm weather and surgical site infections: A meta-analysis. Plast Reconstr Surg Glob Open. 2021;9(7):e3705. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Spencer E, Berry M, Martin P, Rojas-Garcia A, Moonesinghe SR. Seasonality in surgical outcome data: A systematic review and narrative synthesis. Br J Anaesth. 2022;128(2):321. [DOI] [PubMed] [Google Scholar]
- 8.Kane P, Chen C, Post Z, Radcliff K, Orozco F, Ong A. Seasonality of infection rates after total joint arthroplasty. Orthopedics. 2014;37(2):e182. [DOI] [PubMed] [Google Scholar]
- 9.Zalikha AK, Abbas KJ, Karabon P, Hussein IH, El-Othmani MM. The impact of month on joint arthroplasty in-hospital outcomes: The december effect. Arthroplast Today. 2022;16:101. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Rosas S, Ong AC, Buller LT, et al. Season of the year influences infection rates following total hip arthroplasty. World J Orthoped. 2017;8(12):895. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hald JT, Hesselvig AB, Jensen AK, Odgaard A. Revision for periprosthetic joint infection rate stratified by seasonality of operation in a national population of total and unicompartmental knee arthroplasty patients: A register-based analysis. J Bone Jt Infect. 2021;6(5):111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sanders FRK, van’t Hul M, Kistemaker RMG, Schepers T. Seasonal effect on the incidence of post-operative wound complications after trauma-related surgery of the foot, ankle and lower leg. Arch Orthop Trauma Surg. 2020;140(11):1677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ng M, Song S, George J, et al. Associations between seasonal variation and post-operative complications after total hip arthroplasty. Ann Transl Med. 2017;5(Suppl 3):S33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Collins NJ, Roos EM. Patient-reported outcomes for total hip and knee arthroplasty: Commonly used instruments and attributes of a “good” measure. Clin Geriatr Med. 2012;28(3):367. [DOI] [PubMed] [Google Scholar]
- 15.Kingsley C, Patel S. Patient-reported outcome measures and patient-reported experience measures. Bja Education. 2017;17(4):137. [Google Scholar]
- 16.Chow A, Mayer EK, Darzi AW, Athanasiou T. Patient-reported outcome measures: The importance of patient satisfaction in surgery. Surgery. 2009;146(3):435. [DOI] [PubMed] [Google Scholar]
- 17.Koren L, Barak A, Norman D, Sachs O, Peled E. Effect of seasonality, weather and holidays on the incidence of proximal hip fracture. Isr Med Assoc J: Isr Med Assoc J. 2014;16(5):299. [PubMed] [Google Scholar]
- 18.Grønskag AB, Forsmo S, Romundstad P, Langhammer A, Schei B. Incidence and seasonal variation in hip fracture incidence among elderly women in Norway. The HUNT Study. Bone. 2010;46(5):1294. [DOI] [PubMed] [Google Scholar]
- 19.Schlager D, Schwartz JE, Bromet EJ. Seasonal variations of current symptoms in a healthy population. Br J Psychiatry. 1993;163:322. [DOI] [PubMed] [Google Scholar]
- 20.Mersch PP, Middendorp HM, Bouhuys AL, Beersma DG, van den Hoofdakker RH. The prevalence of seasonal affective disorder in The Netherlands: A prospective and retrospective study of seasonal mood variation in the general population. Biol Psychiatr. 1999;45(8):1013. [DOI] [PubMed] [Google Scholar]
- 21.Vajapey SP, McKeon JF, Krueger CA, Spitzer AI. Outcomes of total joint arthroplasty in patients with depression: A systematic review. J Clin Orthop Trauma. 2021;18:187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Leekha S, Diekema DJ, Perencevich EN. Seasonality of staphylococcal infections. Clin Microbiol Infection. 2012;18(10):927. [DOI] [PubMed] [Google Scholar]
- 23.Rana AJ, Sturgeon CM, McGrory BJ, Frazier MV, Babikian GM. The able anterior-based muscle-sparing approach: A safe and effective option for total hip arthroplasty. Arthroplasty Today. 2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Trenberth KE. What are the seasons? Bull Am Meteorol Soc. 1983;64(11):1276. [Google Scholar]
- 25.Yang Z, Ji W, Xia Y, Wang X. Late summer is a risk factor for periprosthetic joint infection after total joint arthroplasty: A retrospective cohort study. Medicine (Baltim). 2023;102(10):e33089. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Khanna A, Betech AA, Chapple AG, Krause PC, Dasa V. Total joint arthroplasty complication and revision surgery rates vary by region and season: Analysis of a large nationwide database. J Am Acad Orthop Surg Glob Res Rev. 2023;7(1):22. [Google Scholar]
- 27.Upperman CR, Parker JD, Akinbami LJ, et al. Exposure to extreme heat events is associated with increased hay fever prevalence among nationally representative sample of US adults: 1997-2013. J Allergy Clin Immunol Pract. 2017;5(2):435. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Engebretsen K, Johansen J, Kezic S, Linneberg A, Thyssen J. The effect of environmental humidity and temperature on skin barrier function and dermatitis. J Eur Acad Dermatol Venereol. 2016;30(2):223. [DOI] [PubMed] [Google Scholar]
- 29.Eccles R. An explanation for the seasonality of acute upper respiratory tract viral infections. Acta Otolaryngol. 2002;122(2):183. [DOI] [PubMed] [Google Scholar]
- 30.Aghdassi SJS, Schwab F, Hoffmann P, Gastmeier P. The association of climatic factors with rates of surgical site infections: 17 Years' data from hospital infection surveillance. Dtsch Arztebl Int. 2019;116(31-32):529. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Fortaleza CMCB, de Oliveira Silva M, Rodrigues FS, da Cunha AR. Impact of weather on the risk of surgical site infections in a tropical area. Am J Infect Control. 2019;47(1):92. [DOI] [PubMed] [Google Scholar]
- 32.Emara AK, Nageeb E, George J, Buttaro MA, Higuera C, Piuzzi NS. Hypovitaminosis D in lower extremity Joint Arthroplasty: a systematic review and meta-analysis. J Orthop. 2020;21:109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.US Department of Commerce, NOAA. National Weather Service. Forecast.weather.gov. https://forecast.weather.gov/MapClick.php?textField1=43.6591&textField2=-70.2568. Accessed November 11, 2023.
- 34.Nakashima D, Kimura D, Watanabe H, et al. Influence of seasonal variations on physical activity in older people living in mountainous agricultural areas. J Rural Med. 2019;14(2):165. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Garriga A, Sempere-Rubio N, Molina-Prados MJ, Faubel R. Impact of seasonality on physical activity: A systematic review. Int J Environ Res Publ Health. 2021;19(1):2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Silva M, Shepherd EF, Jackson WO, Pratt JA, McClung CD, Schmalzried TP. Knee strength after total knee arthroplasty. J Arthroplasty. 2003;18(5):605. [DOI] [PubMed] [Google Scholar]
- 37.Ross L, O'Rourke S, Toland G, et al. Orthopaedic Proceedings. The British Editorial Society of Bone and Joint Surgery, 2022. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Material for The Association of Season of Surgery and Patient Reported Outcomes following Total Hip Arthroplasty by Andrew D. Lachance, Catherine Call, Zachary Radford, Henry Stoddard, Callahan Sturgeon, George Babikian, Adam Rana, and Brian J. McGrory in Progress in Physical Geography