If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Corresponding author: Ivan Z. Liu, BA, Medical College of Georgia Department of Orthopedics, Augusta University, 1120 15th St, BA-3300, Augusta, GA 30912.
To quantify and describe the impact of the COVID-19 pandemic on procedural volume trends in hand and wrist surgery from 2020 to early 2022 at multiple centers.
Methods
In this retrospective comparative study, a real-time, national, federated research database was used to identify patients of interest from 56 health care organizations across the United States. Patients were queried from March 1, 2018, to February 28, 2022. Current Procedural Terminology codes were chosen using the Accreditation Council for Graduate Medical Education’s hand fellowship procedure requirements.
Results
Common hand and wrist surgeries exhibited substantial fluctuations in procedural volume per health care organization during the COVID-19 pandemic. Time periods with considerable procedural volume decreases corresponded with surges in increased COVID-19 caseloads and emergence of COVID-19 variants. Periods of procedural volume increase occurred in the summer of 2020 and immediately following distribution of the COVID-19 vaccine to the public. Fixation of metacarpal fracture, fixation of phalangeal fracture, tendon transfer, flexor tendon repair, and extensor tendon repair consistently showed decreased volumes over the study period. In contrast, ulnar nerve decompression was the only procedure to experience a statistically significant increase in volume over an entire year (2021, +19.2%, P < .001), as compared to before the pandemic.
Conclusions
Major milestones of the COVID-19 pandemic correlated with fluctuations in the number of hand and wrist procedures performed across the United States. Future studies should seek to evaluate the impact of patient backlogs and individual procedure fluctuations on financial impacts, patient outcomes, and orthopedic trainee experience.
In March 2020, health care systems across the United States were forced to rapidly divert resources as the COVID-19 outbreak developed into a pandemic. As a result, elective and nonurgent surgeries were largely canceled.
A set of guidelines released by the American Academy of Orthopaedic Surgeons in April 2020 encouraged the delay of nonurgent and elective surgeries but endorsed prompt management of emergent surgical cases associated with increased morbidity.
Surgical acuity stratification, availability of personal protective equipment, intensive care unit beds, and burden of COVID-19 at a facility were the important factors in case cancelations.
Although stay-at-home orders were instated and employment rates fell, requirements for hand and wrist trauma surgeries remained high, potentially resulting in an unmet demand for hand and wrist procedures.
To our knowledge, no prior studies have quantified the national impact of COVID-19 on hand and wrist surgery procedural volume in the United States. The purpose of this study was to quantify and describe the impact of the COVID-19 pandemic on procedural volume trends in hand and wrist surgery from 2020 to early 2022 at multiple centers. We hypothesized that major milestones of the COVID-19 pandemic would correlate with significant fluctuations in the number of hand and wrist procedures performed.
Materials and Methods
In this retrospective comparative study, TriNetX, a real-time, federated research database, was used to identify patients of interest. TriNetX contains aggregate, deidentified patient information from around the world and consists of multiple subdatabases, with individual health care organizations self-reporting data to the TriNetX network. Most TriNetX data are imported directly from each health care organization’s primary electronic health records system, with additional patient information extracted from clinical documents using the natural language processing and oncology data augmented by information from cancer registries. TriNetX has been verifiably used in current clinical research literature.
Because the data are deidentified and aggregate, the database allows for real-time analysis while maintaining Health Insurance Portability and Accountability Act and General Data Protection Regulation compliance, thereby avoiding the lengthy delays often associated with insurance payor databases. At the time of analysis, the Research Network subdatabase consisted of >80 million patients from 56 health care organizations (HCO) across the United States. Current Procedural Terminology (CPT) codes were chosen using the Accreditation Council for Graduate Medical Education’s hand fellowship procedure requirements, as listed in Table 1.
The database was then queried for all patients undergoing each of the chosen hand and wrist orthopedic procedure types from March 1, 2018, to February 28, 2022. Monthly case volumes and the corresponding number of reporting HCOs for each procedure type were collected for analysis. To account for potential procedure-dependent discrepancies in the number of contributing HCOs, the total monthly case volume was divided by the number of HCOs reporting per month. This generated a mean procedural volume per HCO for each procedure type in each month. Then, these monthly CPT-specific volumes per HCO were averaged chronologically in 3-month time period. This allowed for calculating the seasonal mean procedural volume per HCO for each procedure type for additional analysis. For the control group, operative volumes from 2018 and 2019 were combined to decrease the variance when calculating prepandemic monthly mean procedural volumes for each procedure type.
Pandemic seasons were defined as follows, beginning in March 2020 with the World Health Organization’s declaration of a pandemic: spring 2020 (March−May 2020), summer 2020 (June−August 2020), fall 2020 (September−November 2020), winter 2020 (December 2020−February 2021), spring 2021 (March−May 2021), summer 2021 (June−August 2021), fall 2021 (September−November 2021), and winter 2021 (December 2021−February 2022). The 4 prepandemic seasons were defined as follows: spring 2018/2019 (March−May 2018 + March−May 2019), summer 2018/2019 (June−August 2018 + June−August 2019), fall 2018/2019 (September−November 2018 + September−November 2019), and winter 2019/2019 (December 2018 to February 2019 + December 2019 to February 2020). Finally, these seasonal means were further averaged over 4 consecutive seasons to calculate the mean monthly procedural volumes over each pandemic year and were defined as follows: pandemic 2020 (spring 2020 to winter 2020), pandemic 2021 (spring 2021 to winter 2021), and total pandemic period (spring 2020 to winter 2021). The mean procedural volumes over these pandemic years were then compared with those calculated for the prepandemic years. Descriptive analyses were performed, and comparisons were made using the Student t test. Statistical analysis was conducted using the R statistical software (v4.1.2; R Core Team 2021), with a predetermined level of significance set at P < .05.
Results
Comprising 56 unique HCOs, 381,046 hand and wrist procedures were performed from March 2018 to February 2022 and were included in the analysis. Figure 1 illustrates an overlay of the gross monthly sum of all procedures performed per HCO on daily COVID-19 caseload, as reported by the Centers for Disease Control and Prevention. Notably, there were 4 unique timeframes when procedural volumes per HCO visibly decreased: during the onset of COVID-19 (spring 2020), the first major surge of COVID-19 caseloads (fall 2020 and winter 2020), the onset of the Delta B.1.617.2 variant (summer and fall 2021), and the largest spike in COVID-19 caseloads with the Omicron B.1.1.529 variant (winter 2021) (Fig. 1). Additionally, 2 visible increases in procedural volumes per HCO occurred: immediately following the first few months of the pandemic (summer 2020) and coinciding with COVID-19 vaccine availability (winter 2020) and after 100 million vaccine doses were administered in the United States (spring 2021).
Figure 1Procedural trends during the COVID-19 pandemic. The figure shows an overlay of daily COVID-19 caseloads per day (bar graph, left-hand y axis), average total procedure volume per month per HCO (line graph, right-hand y axis), and notable COVID-19 milestones (vertical annotations). The x axis illustrates chronology, whereas the red and green shading indicate time pre- and postvaccine approval.
The top 5 procedures with the highest volume over the entire review period were as follows: carpal tunnel decompression (97,634 patients), soft tissue reconstruction (79,523 patients), nerve repair (30,242 patients), open reduction and internal fixation/closed reduction and external fixation distal radius fractures (28,584 patients), and flexor tendon repair (23,889 patients) (Table 1). These 5 procedures remained the highest in total procedural volumes in the early pandemic seasons of spring 2020 and summer 2020 (Table 2). However, for all other subsequent seasons of the pandemic, ulnar nerve decompression had a greater procedural volume than flexor tendon repair. Carpal tunnel decompression had the highest procedural volume in every season over the entire review period.
When comparing the total pandemic period to the prepandemic period, only tendon transfer and fixation of metacarpal fracture experienced statistically significant decreases (Table 3). When isolated from the pandemic 2020 year and compared with the prepandemic period, the pandemic 2021 year also experienced a statistically significant decrease in wrist arthroscopy and a statistically significant increase in ulnar nerve decompression compared with the prepandemic timeframe (Table 3).
Table 3Mean Procedural Volumes per HCO by Pandemic Year
Across all 8 pandemic seasons analyzed, spring 2020 experienced the highest number of procedures with significantly different mean procedural volumes per HCO, compared with the prepandemic timeframe, with 10 different procedures experiencing statistically significant decreases. Spring 2021, summer 2021, and winter 2021 experienced the greatest number of procedures with statistically significant increases, compared with the prepandemic seasons with 3 procedure types each. More detailed breakdowns for each season are described as follows.
Spring 2020
Nearly all procedures experienced a decrease in the average seasonal volume per HCO compared with spring 2018/2019 (Table 2). Ten procedures experienced statistically significant decreases in procedural volume. Only wrist arthroscopy and amputation experienced an increase in procedural volume; however, neither procedure reached statistical significance. All other procedures analyzed experienced a decrease in the average seasonal volume per HCO but did not attain statistical significance.
Summer 2020
During summer 2020, 13 of 18 procedures experienced an increase in the average seasonal volume per HCO compared with summer 2018/2019. Thumb carpometacarpal reconstruction (7.3, +17.4%, P < .05) was the only procedure to experience a statistically significant increase in procedural volume in summer 2020. None of the 5 procedures that experienced a decrease in the average seasonal volume were found to be statistically significant compared with prior years.
Fall 2020
In fall 2020, 13 of 18 procedures experienced an increase in the average seasonal volume per HCO compared with fall 2018/2019. Vascular repair (+7.7, 18.2%, P < .05) and ulnar nerve decompression (15.0, +17.6%, P < .05) were the only 2 procedures to experience a statistically significant increase in fall 2020. The remaining 5 procedures experienced decreases compared with the prepandemic volumes but did not attain statistical significance.
Winter 2020
In this season, 9 procedures experienced increases in the average volume per HCO compared with winter 2018/2019, and 9 procedures experienced decreases. The only procedure to experience a statistically significant difference was tendon transfer, which registered a decrease in volume (3.4, −12.3%, P < .05).
Spring 2021
In spring 2021, 14 of 18 procedures experienced an increase in the average seasonal volume per HCO compared with spring 2018/2019. Carpal tunnel decompression (64.3, +17.2%, P < .01), finger arthrodesis or arthroplasty (4.9, +19.4%, P < .05), and ulnar nerve decompression (16.3, +34.7%, P < .001) experienced statistically significant increases. All other procedures did not experience statistically significant changes in the average seasonal procedural volume per HCO.
Summer 2021
In total, 10 procedures experienced an increase in the average seasonal volume per HCO, whereas 8 procedures experienced a decrease compared with summer 2018/2019. Thumb carpometacarpal reconstruction (7.3, +20.2%, P < .05), ulnar nerve decompression (14.2, +17.5%, P < .005), and wrist arthrodesis (2.1, +14.8%, P < .05) experienced statistically significant increases. Extensor tendon repair (6.4, −12.3%, P < .05) and wrist arthroscopy (2.9, −21.0%, P < .01) experienced statistically significant decreases. All other procedures did not experience statistically significant changes in the average seasonal procedural volume per HCO.
Fall 2021
In fall 2021, 6 procedures experienced increases in the average seasonal volume per HCO compared with fall 2018/2019, whereas 11 procedures experienced decreases. Soft tissue reconstruction (48.3, +11.4%, P < .05), ulnar nerve decompression (15.7, +17.4%, P < .05), and vascular repair (7.7, +23.1%, P < .05) experienced statistically significant increases. All other procedures did not experience statistically significant changes in the average seasonal procedural volume per HCO.
Winter 2021
In winter 2021, all procedures except for ulnar nerve decompression experienced a decrease in the average seasonal volume per HCO compared with winter 2018/2019. Fixation of metacarpal fracture (5.6, −27.2%, P < .01) was the only procedure to have a statistically significant difference. All other procedures did not experience statistically significant changes in the average seasonal procedural volume per HCO.
Discussion
This study represents a multi-institutional quantification of the impact of COVID-19 on hand and wrist surgery procedural volume. Our findings of substantial fluctuations in common hand and wrist procedure volumes per health care organization are consistent with our initial hypothesis that major milestones of the COVID-19 pandemic significantly impacted the number of hand and wrist procedures performed. Notably, this analysis suggests that the total monthly procedure volumes per health care organization exhibited an inverse relationship with COVID-19 caseloads and emergence of Delta B.1.617.2 and Omicron B.1.1.529 variants. An increase in the total procedure volume per health care organization occurred in summer 2020 and after distribution of the vaccine to the public. Individual procedures fluctuated substantially with each season of the pandemic. Fixation of metacarpal fracture, fixation of phalangeal fracture, tendon transfer, flexor tendon repair, and extensor tendon repair consistently showed decreased volumes (Fig. 2). In contrast, ulnar nerve decompression was the only procedure to experience a statistically significant increase in volume over an entire year (2021, +19.2%, P < .001) as compared to before the pandemic (Fig. 2). Taken altogether, these COVID-19–related procedural volume fluctuations may have substantial clinical implications related to shifts in patient behaviors, longer-lasting patient backlogs, and negative impacts on orthopedic training.
Figure 2Selected seasonal procedure volumes. The figure shows a selection of individual procedures that consistently experienced either statistical decreases or increases in the mean procedural volume per HCO per pandemic season over the entire study period. Individual procedures that experienced both statistical increases and decreases in multiple successive seasons were not selected. Solid lines in each color represent the mean seasonal procedural volume per HCO during the pandemic, whereas the dotted lines of the same color represent the prepandemic mean seasonal procedural volume per HCO.
Figure 1 offers a graphic representation that summarizes the trends in total common hand and wrist procedure volumes during the COVID-19 pandemic. Previous literature describing procedural volume trends in other elective-heavy surgical subspecialties has shown that total procedure volume decreased in spring 2020 and increased in summer 2020 and spring 2021.
Existing literature analyzing the impact of the COVID-19 pandemic on overall surgical procedural volumes also found a significant decrease in spring 2020 and summer 2020 and that elective procedures experienced more profound decreases both in winter 2020 and throughout the pandemic.
Existing literature is thereby consistent with our findings, suggesting that hand and wrist procedural volume trends are more susceptible and sensitive to variations in COVID-19 caseloads.
Individual procedure volumes fluctuated considerably during the COVID-19 pandemic, with consistently decreased seasonal volumes observed in work-related accidents of fixation of metacarpal fracture, fixation of phalangeal fracture, tendon transfer, flexor tendon repair, and extensor tendon repair. These findings of decreased seasonal volumes are consistent with those reported in the existing literature. A French-based analysis on the incidence of hand trauma during the COVID-19 pandemic also reported a decrease in injuries secondary to work accidents and an increase in neurovascular injuries.
A recent study by Harvard Medical School similarly found a decrease in both elective and urgent hand procedures, with a moderately decreased effect on urgent procedures.
One possible explanation for our finding of decreased work-related surgeries could be substantial increases in unemployment during the pandemic, with unemployment rates reported by the US Bureau of Labor Statistics as increasing from 3.5% in February 2020 to 14.7% in April 2020 and gradually declining to 3.8% by February 2022.
Contrary to other procedures, there was a statistically significant increase in the seasonal volume for ulnar nerve decompression during most of the pandemic seasons compared with the prepandemic levels. This discrepancy could be explained by complicated, individual changes in leisure activity preference during the pandemic. For example, an Oregon State University study found that home-based and digital/online activities increased in frequency, whereas physical and nature-based activities decreased.
It is possible that increased ulnar nerve entrapment and, thereby, decompression surgery is just 1 manifestation of such shifts in interests and hobbies during the pandemic. Nonetheless, with changing incidence rates and evolving patient demands, hand surgeons and their health care teams should remain flexible to best anticipate specific operation-type burdens to maximize patient outcomes.
Our analysis further suggests that after the initial decline in total procedure volume in spring 2020, there followed an increase in total procedure volumes during summer 2020 and fall 2020, sometimes to levels higher than those in the years before the pandemic in summer 2018/2019 and fall 2018/2019, respectively. The creation of these initial surgical backlogs with ensuing catchup has been well-documented on smaller scales in prior literature. However, it remains equivocal whether the magnitude of these catchups sufficiently compensated for the size in backlog. For example, a recent Hospital for Special Surgery analysis found that their annual surgery volume in 2020 reduced by 20.2% but never sufficiently compensated for the setback experienced during the lockdown, with other studies reporting similar findings.
Our findings further support this existing literature that reported a considerable orthopedic backlog and subsequent rebound on a national scale; however, because of database limitations, we were unable to assess whether this rebound was adequate in counteracting the initial spring 2020 decrease. A novel method to estimate patient demand for elective hand surgery using Google Trends suggests that the interest in elective procedures declined in spring 2020 and subsequently rebounded and remained in steady state thereafter but did not surpass the prepandemic levels.
Although it is important to note that Google Trends fails to account for what percentage of interested patients actually receive operative care, the lack of evidence for this backlogged demand in elective surgery implies that other factors may have played a role in perpetuating this decreasing interest even after the initial pandemic seasons. Some possible explanations include patients’ fear of exposure secondary to surgery during the pandemic, unemployment-related loss of insurance, and financial stress. In total, this well-documented backlog and possibly inadequate rebound of elective procedures not only poses financial burdens on already strained hospital systems but also negatively impacts hand and wrist patient outcomes with probable extensive delays to care.
Finally, our analysis of individual procedure types showed considerable variation in the average seasonal volume per health care organization, although with seemingly minor magnitudes of calculated change per HCO. There ostensibly exists a question of the clinical significance of these calculated changes. Prior literature has suggested that the impact of these variations has resulted, at the least, in meaningful detriments to orthopedic training. A Northwestern University study reported that 59% of hand surgery fellows and fellowship directors across the United States experienced >75% decrease in case volume, with a 52% decrease in mean hours worked.
Stochastic simulation estimates on the raw scale of orthopedic case volume decline suggest that, optimistically, >1 million patients await in backlog by spring 2022.
Given the self-reported impact by orthopedic trainees and the sheer size of case volume lost, our findings of hugely variable changes in individual procedure volume provide additional context to the negative impact of COVID-19 on orthopedic learning and experience. Additionally, it is important to contextualize our findings of individual mean procedural volumes by considering that they are all calculated by dividing the pooled observed procedure counts in 1 month by up to 50+ unique HCOs, averaged over a season or year. In this context, a seasonal mean decrease of 1.2 procedures per HCO for fixation of metacarpal fractures over spring 2020 compared with spring 2018/2019 (Table 2) represents a statistically significant change of approximately 3.6 fewer procedures per HCO across the season observed over a nationally representative sample of 50+ HCOs. Because TriNetX does not allow for further discrimination of each HCO type (ie, community hospital, major academic center) for privacy compliance, HCOs with more robust hand and wrist surgery centers than others may have experienced even greater real-world changes in procedural volume than our calculated findings.
Possible limitations of this study include those inherent to databases and retrospective studies. These potential limitations include the possibility of errors in database coding and data entry, as well as potential losses in patient follow-up not noted in the database. Because CPT and International Classification of Diseases, Tenth Revision codes were used for patient identification, the study lacks full clinical context regarding pertinent patient history and treatment course. For example, disease severity at presentation, urgent versus nonurgent status, and preoperative health measures are directly correlated with preoperative approval and wait-time duration. Additionally, TriNetX contains deidentified data presented on a national level and, thereby, lacks sensitivity to geographical bias stemming from local surges in COVID-19 cases and unique institutional, municipal, or state-wide differences in elective operation guidelines. Further TriNetX-specific database limitations include an inability to further include/exclude data-reporting HCOs (eg, including only HCOs that have an established hand and wrist surgery center) and that data sets containing unique and anonymous patient identifiers are pay-walled, making multivariate analyses financially difficult to perform. Despite these limitations, we believe that the results from our larger cohort sizes, continuously updated patient information, and nationally representative study population consisting of multiple surgeons, academic centers, and community hospitals provide valuable clarification on the impact of COVID-19 on the number of hand and wrist procedures performed across the United States.
References
Guy D.
Bosco J.
Savoie F.
AAOS Guidelines for Elective Surgery During the COVID-19 Pandemic.
Declaration of interests: No benefits in any form have been received or will be received related directly or indirectly to the subject of this article.
00194-3&id=gr1.jpg){kind=link}
00194-3&id=gr2.jpg){kind=link}