A look at the global impact of COVID-19 pandemic on neurosurgical services and residency training

Abstract

Background

The COVID-19 pandemic has left an indelible effect on healthcare delivery and education system, including residency training. Particularly, neurosurgical departments worldwide had to adapt their operating model to the constantly changing pandemic landscape. This review aimed to quantify the reduction in neurosurgical operative volume and describe the impact of these trends on neurosurgical residency training.

Methods

We performed a comprehensive search of PubMed and EMBASE between December 2019 and October 2022 to identify studies comparing pre-pandemic and pandemic neurosurgical caseloads as well as articles detailing the impact of COVID-19 on neurosurgery residency training. Statistical analysis of quantitative data was presented as pooled odds ratio (OR) and 95% confidence intervals (CI).

Results

A total of 49 studies met the inclusion criteria, of which 12 (24.5%) were survey-based. The case volume of elective surgeries and non-elective procedures decreased by 70.4% (OR=0.296, 95%CI 0.210–0.418) and 68.2% (OR=0.318, 95%CI 0.193–0.525), respectively. A significant decrease was also observed in functional (OR=0.542, 95%CI 0.394–0.746), spine (OR=0.545, 95%CI 0.409–0.725), and skull base surgery (OR=0.545, 95%CI 0.409–0.725), whereas the caseloads for tumor (OR=1.029, 95%CI 0.838–1.263), trauma (OR=1.021, 95%CI 0.846–1.232), vascular (OR=1.001, 95%CI 0.870–1.152), and pediatric neurosurgery (OR=0.589, 95%CI 0.344–1.010) remained relatively the same between pre-pandemic and pandemic periods. The reduction in caseloads had caused concerns among residents and program directors in regard to the diminished clinical exposure, financial constraints, and mental well-being. Some positives highlighted were rapid adaptation to virtual educational platforms and increasing time for self-learning and research activities.

Conclusion

While COVID-19 has brought about significant disruptions in neurosurgical practice and training, this unprecedented challenge has opened the door for technological advances and collaboration that broaden the accessibility of resources and reduce the worldwide gap in neurosurgical education.

Introduction

COVID-19, the disease caused by SARS-CoV-2, has been characterized by the World Health Organization as a public health emergency and later declared as global pandemic [1]. To date, more than 625 million people worldwide have tested positive for COVID-19 and at least 6.5 million deaths have been attributed to it [2]. Resources required to care for this patient population and the overwhelming demands have strained the healthcare systems rapidly in many countries [3]. The prioritization of surgical services has also shifted: non-urgent elective surgical cases and outpatient clinics were placed on hold, while emergent cases continued but with revised management protocols [4], [5], [6]. Within neurosurgical care, many procedures are time-sensitive or semi-urgent – treating sooner may prevent long-lasting neurological deficits and potentially benefit more from surgical intervention [7]. Triaging non-elective neurosurgical indications, therefore, presented an unprecedented challenge under the constantly changing pandemic landscape that required meticulous weighing of the risks and benefits [8].

In addition to the impact on case volume, quality of neurosurgical residency training is another aspect that has been negatively influenced by the COVID-19 pandemic [9]. Many programs underwent temporary restructuring of their services to both minimize the number of residents exposed to COVID-19 as well as to maintain workflow to allow urgent reallocation of hospital resources [10]. While the training time for neurosurgical residency is long (ranging from 5 to 8 years), even three months of reduction in surgical activities and clinical experiences may have far-reaching effects on training [9]. Adaptive measures reported to supplement resident training included the use of webinar-type online platforms for conferences, at-home microsurgical skill training programs, surgical stimulation sessions, and research activities that can be performed remotely [11], [12], [13], [14].

Prior to the COVID-19 pandemic, disparity in neurosurgical training between high-income countries (HIC) and low- and middle-income countries (LMICs) was already a prominent problem leading to maldistribution of neurosurgeons and neurosurgical care [15]. The outbreak of COVID-19 further deepened the existing disparities [16]. For instance, despite the abovementioned innovative platforms for neurosurgical training and education can be distributed rapidly across the world, LMICs would have limited ability to implement those novel technologies [13,14]. There has also been a reduction in international training opportunities, including observership programs, elective rotations, clinical and research fellowship positions, and in-person conferences [17].

Understanding the impact of the COVID-19 pandemic on neurosurgical training can provide insight for program directors and faculty members to discern which aspects of residency training are most vulnerable to changes and to recognize what modifications can help to maintain consistency in training quality as well as to help guide response to future pandemics. In this study, we sought to quantify the reduction in neurosurgical case volume and to ascertain the associated effect on residency training worldwide through a systematic review of relevant published articles and further analysis of the results.

Methods

The manuscript was prepared in accordance with the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines [18]. Search strategy, selection criteria, and outcome measures were determined in advance.

Search strategy

A comprehensive systematic search was conducted through PubMed (National Library of Medicine) and EMBASE (Elsevier) from December 2019 to September 2022. Terms synonymous with “COVID-19”, “neurosurgery” and “residency training” were utilized with Boolean operators to maintain high inclusivity for the initial search. We also performed manual searches of the bibliographies to identify relevant studies. The search was limited to English language publications and human subjects. Full search strategy can be found in Supplementary Table 1.

Selection criteria

The search results were entered into Rayyan (https://www.rayyan.ai/) (Rayyan Systems Inc, Cambridge, Massachusetts, USA), which is a web mobile application for systematic review [19]. Two authors independently screened titles and abstracts of all identified studies with a third author who provided an assessment in case of any disagreement. We included studies that directly compared neurosurgical case volume in pre-COVID-19 era and COVID-19 era as well as articles detailing the impact of COVID-19 on neurosurgical residency training worldwide. Studies were excluded if they did not provide comparative data, compared the wrong time periods, focused on methodological or technological innovations rather than residency training, and did not provide separate data from other surgical specialties. Conference abstracts, commentaries, and cadaver or animal studies were also excluded. Of note, we included letters to the editor and editorials as most of the neurosurgery journals were accepting studies as only editorials during the COVID-19 pandemic.

Data collection and quality assessment

Following the finalization of the included studies, relevant data including first author, title, publication date, country of the enrolled population, study design, and pandemic period reported were extracted into a standardized, pilot-tested template. The primary outcome of interest was the impact on neurosurgical residency training, which could be measured in terms of caseloads, redeployment, work hours, research activities, change in educational sessions (e.g. morbidity and mortality conference, grand rounds, and board review), and solution implemented to cope with these difficulties. The quality of included cohort or case control studies was evaluated using the 9-star Newcastle-Ottawa Quality Assessment Scale (NOS) [20].

Statistical analysis

Studies with data available for case volumes in the pre-COVID-19 era and COVID-19 era were included in the quantitative meta-analysis. Effect sizes were presented as odds ratio (OR) with their corresponding 95% confidence interval (CI) and displayed in the form of a forest plot. We assessed heterogeneity across individual studies by inspecting forest plots and with I² index and Cochran's Q statistical test. If I² was >50% and p-value <0.05, a random-effects model (DerSimonian and Laird method) was chosen; otherwise, the fixed-effect model was used to pool the estimate. Potential publication bias was evaluated using the Eggers’ regression asymmetry test. We considered a two-sided p-value <0.05 statistically significant. All statistical analyses were performed using RStudio (https://www.r-project.org/) (RStudio, Boston, Massachusetts, USA).

Results

Study identification

A systematic search of the literature yielded a total of 606 and 628 articles from PubMed and Embase, respectively. After removing duplicates, 935 articles were left; their titles and abstracts were screened, and 78 relevant articles were identified for full-text review. Overall, this study collated data and information from 49 articles [11,12,17,[21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54], [55], [56], [57], [58], [59], [60], [61], [62], [63], [64], [65]], with 25 of them included in the quantitative synthesis (Fig. 1 ). Twenty-two of these articles were retrospective in nature, two had data collected both retrospectively and prospectively, one was both a retrospective cohort study and a survey. The remaining 14 articles were surveys and were included in the qualitative synthesis. By country involved, most studies enrolled patients from the United States (n=21), followed by India (n=4), Egypt (n=4), United Kingdom (n=3), and Spain (n=3). For quality assessment, the mean Newcastle–Ottawa scale was 6.2±0.9 (range, 5 to 8), suggesting that majority of the included cohort studies were of moderate methodological quality. Study characteristics of the included articles are summarized in Table 1 .

Fig. 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram showing the number of articles identified and excluded at each stage of the literature search.

Table 1.

Descriptive summary of included studies in systematic review and meta-analysis.

Author, Year	Country Involved	Study design	Pandemic Period Reported	Impacts on Residency Training	NOS
Alhaj et al. 2020 [32]	Kuwait, Canada, USA, Saudi Arabia, Italy, Serbia	Cross-sectional survey	N/A	98% felt training was affected 90% felt mental health was affected 100% reported their social life was affected	N/A
Aljuboori et al. 2021 [12]	USA	Retrospective cohort and survey	Mar 2020 – Apr 2020	87.5% stated didactics moved to online only 37.5% felt surgical skills was negatively affected 87.5% reported more time for research activities	7
Almufarriji et al. 2021 [35]	Saudi Arabia	Survey	N/A	58.8% reported stopping all educational activities 35.8% increase in on-call duties 9.5% redeployed	N/A
Arnaout et al. 2020 [56]	USA	Letter to the editor	N/A	Actively reduced elective surgical cases and in-person visits Restructured services: residents contributed to daily patient care remotely, implemented a coverage model similar to a weekend coverage Eliminated “double” scrubbing by senior residents and junior residents Moved educational activities virtually 50% redeployed	N/A
Ashkan et al. 2021 [27]	United Kingdom	Retrospective and Prospective cohort	Jan 18, 2020 – May 15, 2020	55.6% reduction in overall operative volume 31.5% increase in proportion of emergency operations	8
Ashry et al. 2020 [21]	Egypt	Survey	N/A	56% reported reduction of actual role in surgical steps 10% redeployed 88% stated transition to virtual educational methods 100% felt suffering from financial strains 68% reported burnout symptoms Significant increase in research hours	N/A
Azab et al. 2021 [37]	Egypt	Retrospective cohort	Mar 2020 – Dec 2020	Decrease in mean operative time (pre: 98±8.3 vs. during: 94.3±11.4 min)	6
Bambakidis et al. 2020 [57]	USA	Editorial	N/A	Decreased resident staffing to 50% of normal (1-week on and 1-week off model) Significant drop in elective and nonessential surgical volume Limited surgical cases to a single resident to preserve PPE Transitioned to virtual conferences	N/A
Bajunaid et al. 2020 [38]	Saudi Arabia	Retrospective cohort	Mar 31, 2020 – May 20, 2020	44% reduction in overall operative volume	6
Benner et al. 2022 [41]	USA	Retrospective cohort	Apr 1, 2020 – Apr 30, 2020	58.5% reduction in overall operative volume	6
Burks et al. 2020 [31]	USA	Retrospective cohort	Apr 1, 2020 – Apr 30, 2020	Operative case totals were lower at all levels of training Significant decrease in the number of residents participating in each case (p<0.01) Significant decrease in cases compared to the year before (p<0.01)	7
Bray et al. 2020 [58]	USA	Letter to the editor	N/A	Held weekly online “town-hall meetings” Service subdivided to 2 working teams (2-week cycle) Transitioned to telemedicine for all outpatient visits Redeployed to neurological Critical Care team	N/A
Cerda-Vargas et al. 2021 [42]	Spain, Mexico, Argentina, Brazil	Cross-sectional survey	N/A	66.2% felt training was affected 86.3% felt either or both of their physical and mental health were affected 20.1% reported increase in weekly academic hours	N/A
Cheserem et al. 2020 [43]	Egypt, Zimbabwe, South Africa	Cross-sectional survey	Apr 16, 2020 – May 21, 2020	Marked reduction in clinical activities (-80% of elective surgery, -83% of clinics, -38.5% of emergency surgery) 61.0% reported receiving online teaching 57.72% had examinations postponed and 19.51% had examinations cancelled 11.38% redeployed 32.52% reported training rotation suspended 23.58% stated not receive a formal salary	N/A
Dash et al. 2020 [44]	India	Cross-sectional survey	May 7, 2020 – May 16, 2020	67.5% reduction in the average number of surgeries performed per month 32.6% reduction in the number of academic sessions 61.02% reported shift to online videoconferencing sessions 13.56% had academic sessions stopped 88.14% felt training in terms of clinical and operative skills were adversely impacted 53.39% feared being less competent 48.30% felt increase in work-related stress	N/A
Deora et al. 2020 [45]	India	Retrospective cohort	Mar 1, 2020 – Jul 31, 2020	57% reduction in overall operative volume	5
ElGhamry et al. 2021 [46]	United Kingdom	Retrospective cohort	Mar 20. 2020 – Aug 31, 2020	Significant reduction in referral and operative caseloads (p<0.001)	7
Eichberg et al. 2020 [16]	USA	Letter to the editor	N/A	Halted all elective cases but continue to schedule urgent and emergent cases 75% diminished in overall surgical volume Implemented telehealth technology integrated into electronic medical record All in-person conferences have been replaced by video teleconferences Devised rotating resident schedules to minimize viral exposure and burnout	N/A
EI-Ghandour et al. 2020 [59]	International (96 countries)	Letter to the editor	Mar 20, 2020 – Apr 3, 2020	71.4% reported decreased workload 62.5% reported financial burden 26.7% reported cessation of research activities	N/A
Galarza et al. 2020 [60]	Spain	Letter to the editor	N/A	Redeployed to serve on COVID-19 teams Reduced to 27 to 35 surgical cases per day	N/A
Goyal et al. 2020 [47]	India	Retrospective cohort	Mar 25, 2020 – May 31, 2020	52.2% reduction in overall operative volume	6
Jean et al. 2020 [62]	International (60 countries)	Survey	N/A	46.1% reported operative volume dropped more than 50%	N/A
Khalafallah et al. 2020 [26]	USA	Retrospective cohort	Mar 4, 2020 – Apr 17, 2020	97.12% reduction in in-person clinic visits 44.68% reduction in inpatient census 15.4% redeployed Transition of grand rounds and M&M to online videoconference 50% increase in educational session attendance 75% felt didactics were negatively affected 87.5% reported didactics transited to online only	8
Khalafallah et al. 2020 [22]	USA	Survey	N/A	74.8% reported decreased weekly work hours 67.6% reported decreased ability to meet ACGME operative case minimums 15.3% redeployed 82.0% participated in remote didactic lectures 66.7% reported increase in research activities	N/A
Khan et al. 2021 [24]	USA	Retrospective cohort	Mar 16, 2020 – May 8, 2020	Decreased resident coverage to a 1 week on and 1 week off model Decreased operative volume in spine cases and functional cases (p<0.001)	6
Kilgore et al. 2021 [30]	USA	Retrospective cohort	Mar 2020 – June 2020	77% reduction in neurosurgical volume Significant increased incidence of multiple residents scrubbing the same case (p=0.011)	6
Laskay et al. 2020 [61]	USA	Letter to the editor	N/A	Restructured resident coverage to 7-day cycle Department issued an official requirement for all nonessential cases to be cancelled Initiated teleconsultation program for trauma patients with nonurgent or nonoperative pathology Limited in-person conference attendance to 10 individuals Utilized educational resources offered by the Congress of Neurological Surgeons	N/A
Lubansu et al. 2020 [48]	Belgium	Retrospective cohort	Mar 6, 2020 –May 10, 2020	50% reduction in overall operative volume	6
Meybodi et al. 2020 [49]	Iran	Retrospective cohort	Mar 2020 – Jun 2020	31% reduction in overall surgical cases Considerable reduction in subspecialized educational surgeries	6
Nabil et al. 2022 [50]	Egypt	Retrospective cohort	Jan 2020 – Jun 2020	38% reduction in overall operative volume	5
Pannullo et al. 2020 [63]	USA	Letter to the editor	N/A	Cancelled all elective surgeries 65.6% reduction in operative volume 50% clinic visits moved to telehealth 42.9% redeployed Halted all basic and translational experiments Transited to virtual grand rounds, M&M conferences, resident journal club, and department webinars	N/A
Patel et al. 2020 [51]	USA	Retrospective cohort	Mar 23, 2020 –May 8, 2020	40% reduction in weekly procedural volume 28% decrease in weekly neurosurgical consultation 47% decrease in outpatient clinic encounters	6
Pelargos et al. 2020 [11]	USA, Canada	Survey	Apr 17, 2020 – Apr 30, 2020	35.1% providing non-specialty care to COVID-19 patients 82% reduction in inpatient and outpatient volumes 91% reported decrease in work responsibility, with significant decrease in work hours (p<0.0001) 33.7% felt residency education was negatively affected, with senior trainees more likely to be concerned 26.5% concerned limit ability to get desired employment or fellowship Significant increase in number of trainees spending >4 h on didactics p<0.0001	N/A
Petr et al. 2022 [52]	Austria, the Czech Republic	Retrospective cohort	Jan 2020 – Dec 2021	Reduction in number of traumatic brain injuries, spine conditions, and chronic subdural hematomas	5
Saad et al. 2020 [28]	USA	Retrospective cohort	Mar 16, 2020 –Apr 15, 2020	80% reduction in case volume 59% decrease in numbers of bedside procedures 20% reduction in occupancy of neurointensive care unit bed Transition of M&M to virtual conference	7
Sahin et al. 2021 [53]	Turkey	Survey	N/A	54.8% felt theoretic education was negatively affected 21.3% reported transition to completely online teaching sessions 38.3% had teaching sessions not held at all 78.7% felt surgical training was negatively affected, with decreased case volume (60.9%) being the most cited reason 42.1% felt increased concerns about residency training and future career	N/A
Sarpong et al. 2021 [54]	USA	Retrospective cohort	Mar 8, 2020 – Jun 8, 2020	61.0% reduction in case volume	7
Sudhan et al. 2021 [40]	India	Retrospective cohort	Mar 15, 2020 – Sep 15, 2020	42.75% reduction in caseloads No significant difference in overall incidence of emergency and essential surgeries (p=0.482)	6
Suryaningtyas et al. 2020 [23]	Indonesia	Retrospective cohort	Apr 2020 – Jun 2020	50% reduction in overall case volume Reorganized the residents for service in 10–20 schemes (10-day on service, followed by 20-day off hospital)	7
Tavanaei et al. 2021 [39]	Iran	Retrospective and Prospective case control	Jun 1, 2020 – Sep 1, 2020	30% reduction in overall case volume	6
Theofanis et al. 2020 [64]	USA	Letter to the editor	N/A	Devised a new call schedule with 2 teams Mandated residents who were not working stay home Redeployed to “Line Service in intensive care units	N/A
Tzerefos et al. 2021 [33]	Greece, Switzerland, Spain, United Kingdom	Survey	N/A	88.8% felt education was negatively influenced 92.5% reported a reduction in hands-on surgical exposure 92.5% participated in online educational activity 29.9% redeployed 44.8% reported more time for research work 65.7% reported more time for self-study	N/A
Velnar et al. 2022 [34]	Slovenia	Retrospective cohort	Mar 2020 – Apr 2021	No important decline of the number of operated patients	5
Wali et al. 2020 [36]	USA	Retrospective cohort	Mar 16, 2020 – Jul 6, 2020	Only moderate diminish of daily total cases (pre: 6.9 vs. during: 5.8 cases)	5
Weber et al. 2020 [55]	USA	Letter to the editor	N/A	Transitioned from 12 h call periods to 24 h call to minimize face-to-face sign-out Implemented a daily lecture curriculum (1.5 h every day) led by senior residents Transitioned from pre-COVID resident group fitness to in home exercise routines Established a dedicated weekly meeting in which the Program Director and Chairman run a “Town Hall” session for resident concerns, ideas, and questions	N/A
Weiner et al. 2021 [65]	International	Survey	Mar 27, 2020 – Apr 4, 2020	35.6% respondents identified as “high telehealth users” 80.1% were interested in online education	N/A
Wittayanakorn et al. 2020 [17]	Indonesia, Malaysia, Philippines, Singapore, Thailand	Survey	May 22, 2020 – May 31, 2020	82% reduction in elective operations 62% reduction in emergency operations 76% felt training was significantly impacted Transition to virtual meetings for grand rounds and M&M 33% noted a decrease in research productivity	N/A
Zhang et al. 2022 [29]	USA	Retrospective cohort	Jan 2020 – Dec 2020	58% decrease in caseloads for junior residents and 45% for senior residents	6
Zoia et al. 2020 [25]	Italy	Survey	Mar 2020 – May 2020	72.4% reported reduced time in neurosurgical department 78.6% performed less operations 16.1% performed no operations 55.7% reported increased production of scientific papers	N/A

Abbreviation: M&M, morbidity and mortality; N/A, not applicable; NOS, Newcastle Ottawa scale; USA, United States of America.

Elective and non-elective neurosurgical services

Eight studies reported the case volume of elective neurosurgical procedures performed both before and during the COVID-19 pandemic [27,[37], [38], [39], [40],47,50,54]. Collectively, the number of elective surgeries decreased by 70.4% (OR=0.296, 95% CI 0.210–0.418, p<0.001) (Fig. 2 A). A comparison of non-elective neurosurgical procedures was reported in 9 studies, with an overall decrease of 68.2% (OR=0.318, 95% CI 0.193–0.525, p<0.001) (Fig. 2B) [12,27,[37], [38], [39], [40],[46], [47], [48]]. Significant publication bias was observed for both elective and non-elective case volumes (p=0.034 and p=0.001, respectively).

Fig. 2.

Forest plot for comparison of number of (A) elective and (B) non-elective neurosurgical procedures before and during COVID-19 pandemic. OR, odds ratio; CI, confidence intervals.

Neurosurgical subspecialty experience

Among the neurosurgical subspecialties, functional services experienced the highest reduction in case volume (45.81% reduction; OR=0.542, 95% CI 0.394–0.746, p<0.001; Fig. 3 A) followed by spine cases (45.5% reduction; OR=0.545, 95% CI 0.409–0.725, p<0.001; Fig. 3B), and skull base surgery (30.6% reduction; OR=0.695, 95% CI 0.572–0.845, p<0.001; Fig. 3C). No statistically significance in case volume between pre-pandemic and pandemic era was found for neurosurgical oncology (OR=1.029, 95% CI 0.838–1.263; Fig. 4 A), neuro-trauma (OR=1.021, 95% CI 0.846–1.232, p=0.833; Fig. 4B), neurovascular surgery (OR=1.001, 95% CI 0.870–1.152, p=0.991; Fig. 4C), and pediatric neurosurgery service (OR=0.589, 95% CI 0.344–1.010, p=0.055; Fig. 4D). Except for spine (p=0.449), skull base (p=0.132), neuro-trauma (p=0.471), and neurovascular surgery (p=0.398), potential publication were observed for all other subspecialty pooled estimates (all p<0.05).

Fig. 3.

Forest plot for case volume comparison of (A) functional, (B) spine, and (C) skull base service before and during COVID-19 pandemic. OR, odds ratio; CI, confidence intervals.

Fig. 4.

Forest plot for case volume comparison of (A) neurosurgical oncology, (B) neuro-trauma, (C) neurovascular, and (D) pediatric service before and during COVID-19 pandemic. OR, odds ratio; CI, confidence intervals.

Possible effects of COVID-19 pandemic on neurosurgery residency training

The collated data of impacts on residency training are summarized in brief in Table 1 and the modifications implemented are presented in Table 2 . Majority of the included studies observed a significant decrease in overall caseloads, both surgical and non-surgical, resulting in diminished clinical exposure for residents [12,[21], [22], [23], [24], [25], [26], [27], [28],[30], [31], [32], [33],35,[37], [38], [39], [40], [41], [42], [43], [44], [45], [46], [47], [48], [49], [50], [51], [52], [53], [54],[56], [57], [58], [59], [60], [61], [62], [63], [64], [65]]. Ashry et al. found that 56% of the residents had a reduction in their actual role in the surgical steps and the number of residents scrubbing in the same procedure had been decreased to the minimum for preservation of personal protective equipment (PPE) [21]. A survey completed by El-Ghandour and colleagues revealed that elective surgery/clinic cancellation occurred more frequently in African countries and that theyr were utilizing PPE significantly less commonly in LMICs (OR=0.441, 95% CI 0.216-0.900, p<0.05) [59]. Some studies suggested restructuring their resident rotations to incorporate altered working patterns: Suryaningtyas et al. proposed a 10-day on and 20-day-off model, whereas Khan et al. reported a decrease in the resident coverage to one week-on service and one week-off hospital [23,24]. Fifth studies reported a formal reduction of resident presence, with a significant decrease in working hours and responsibility [21,22,26,[30], [31], [32],35,40,44,50,[56], [57], [58],62,64]. However, some residents experienced increased burdens and anxiety due to the extra work caused by the pandemic, including redeployment to help manage COVID-19 patients or COVID-related problems when needed [12,21,42,44]. Moreover, the reduction in operative volumes and clinical experiences has caused concern amongst residents about meeting their annual training requirements [22,26].

Table 2.

Modifications implemented on neurosurgical services and residency training of each included studies.

Abbreviation: HIC, high income country; LMIC, low-middle income country.

Color code: green signifies that the change has been adopted in the study; red signifies that the change has not been adopted in the study.

Aside from changes to clinical activity, educational and academic pursuits have also been impacted [21,42,56]. There is a rapid increase in popularity of online conferences and telemedicine [58]. Some residency programs took advantage of the various complementary virtual didactic lectures, such as American Association of Neurologic Surgeons webinars, Congress of Neurological Surgeons (CNS) Online Grand Rounds, CNS Virtual Visiting Professor, and ad hoc national and international virtual conferences [11]. For many residents, this transition to digital platforms has improved the access to educational materials [12,57]. It is worth noting that surveys conducted in HICs reported a higher percentage of technological shift in comparison to LMICs (87.5% vs. 61.5%, respectively) [17,26,53]. Furthermore, the decrease in case volumes and working hours at the neurosurgical department allowed more time for preparation of board examination and research activities [17,22,33,59]. In Italy, for instance, 71.9% of responders affirmed increased time spent for studying and 55.7% reported an increase in research productivity [25]. However, in LMIC, 33.0% of residents noted a decline in their research output and 71.0% reported missing out on opportunity for education and training [17].

Discussion

The COVID-19 pandemic has presented an unprecedented challenge for virtually every aspect of healthcare worldwide, with all healthcare systems grappling with limitations in resources, space, and staffing [66]. In an attempt to contain transmission and conserve resources, clinical team underwent temporary restructuring and adopted a new workflow to meet the demands of the pandemic [67]. Residents are an indispensable part of any institution's workforce and, therefore, were the most and first affected by the spread of COVID-19 infections [68]. To the best of the authors’ knowledge, this is the first study of its kind to quantify the global impacts of the COVID-19 pandemic on neurosurgical services and the associated effect on residency training. All selected articles had seen a decrease in operative volume, with some reporting concerns over residents’ ability to meet Accreditation Council for Graduate Medical Education operative case requirements and about future career prospects. Redeployment to non-neurosurgical roles varied across studies, ranging from 0% to 36.0% of the residents [17,26,33,60,63,64]. While these restructurings affected the operative skills and clinical experience, adapting to the new normal provided an opportunity for new technologies and additional time for research and self-directed studying.

Impact on elective and non-elective cases

Perhaps most obviously, the COVID-19 pandemic placed a substantial impact on the operative component of neurosurgical training [22,26]. A steep decline in operative volume was noted during the COVID-19 period in comparison to pre-COVID-19 times, with elective cases decreasing by 70.4% and non-elective surgeries decreased by 68.2%. There were multiple factors that could contribute to the case reduction. First, patients may defer or avoid medical attention due to fear of contracting COVID-19 in healthcare settings [69]. Second, residents and physicians were redeployed to aid frontline workers in managing the overwhelming COVID-19 patient load [70]. The redeployment rate of residents had been noticed to be higher in LMICs, according to a multi-country survey [17]. The lack of personnel in the neurosurgical department forced non-essential surgeries to get postponed or canceled [70]. Moreover, the widespread uncertainty in regard to PPE availability, transmission risk, disease management, and planning to prioritize intensive care unit beds had led the Centers for Medicare & Medicaid Service to place a moratorium on postponing all non-emergent, elective surgeries [71]. These challenges were even more pronounced in LMICs given their limited PPE access and significant shortage of supplies, such as ventilators and test reagents [59] HICs had the financial and diplomatic resources to obtain PPE, medical equipment, diagnostic tests, and healthcare workers at the detriment of LMICs [72]. The reduction in surgical volume, however, was less in LMICs, as they carry a disproportionate share of the global burden of surgical cases that cannot be postponed safely [73]. African countries, for example, contain 15% of the global volume of neurosurgical diseases, but their healthcare facilities have access to less than 1% of neurosurgeons worldwide, making training capable and competent neurosurgeons a fundamental step in achieving equal access [74]. The emergent procedures were also expected to decrease, as traffic volumes, social events, sports activities, and assaults were all decreased due to lockdowns and stay-at-home orders [75].

Impact on neurosurgical subspecialty cases

When analyzing by neurosurgical subspecialties, we observed a pronounced reduction in caseloads for spine, functional, and skull base surgery. These findings are expected given spine, functional, and skull base cases were relatively less likely to be emergent in comparison to neurosurgical oncology, neuro-trauma, neurovascular surgery, and pediatric neurosurgery service [24,27]. The disproportionate decrease in spine, functional, and skull base cases may carry significant implications for both the patients and the trainees. Delays in surgical intervention or adoption of a more conservative approach increases the risk of neurological deterioration, permanent loss of function, and treatment failure in these patient populations [7,28]. For trainees, junior residents were among the most impacted as the bulk of their operative volume was routine elective procedures and learning the basics of operative techniques [24,28,29]. Kilgore et al. also suggested that one may see a rise in the number of residents pursuing fellowships in the coming years as a result of a steep decline in elective cases [30]. In addition, our study found a substantial between-study heterogeneity for the rate of neurosurgical oncology procedures. Patients with tumors, such as cerebellar metastasis or glioblastoma, and presenting symptoms, such as acute hydrocephalus or cerebral herniation, are generally considered a priority and should have surgery scheduled in a timely manner [62,76]. However, for patients with benign or low-grade tumors that are asymptomatic, surgical intervention may be postponed until a safer time [76]. Cessation of majority of the endoscopic endonasal procedures due to concerns of disseminating COVID-19 infection may be another reason leading to the discrepancy in neurosurgical oncology case volume between studies [28].

Negative impact on neurosurgical residency training

Due to a significant reduction in the number of surgical cases, bedside procedures, and outpatient clinic encounters, neurosurgical residents have been hit hard by the COVID-19 pandemic from an educational and training perspective [12,31,77]. A common theme across the included studies was the widespread worry about failing to meet case number requirements and operative competency. Those obstacles were not only faced by neurosurgical residents but also echoed by residents in other specialties [77,78]. Another phenomenon observed was the decline in mental health and increase in burnout symptoms, with as high as 90% of the neurosurgical residents worldwide reported being affected [32]. Some contributing factors may include the fears of being infected and transmitting to family members, little social support due to quarantine, shortage of staff and personal protective equipment, redeployment in haste without adequate training and supervision, altered rotation schedules, and financial strains [21,22,33,79].

Neurosurgery residency program adaptations and solutions

As part of an effort to combat the loss of clinical experience, various assisted teaching methods were suggested. Grand rounds, morbidity and mortality conference, board reviews, and lecture series were provided via digital platform [11,28,63,80]. Some programs initiated streaming of procedures online and utilized virtual reality-based surgical stimulators to practice psychomotor surgical techniques [81,82]. However, these immersive technologies were often not available in Sub-Saharan Africa and other LMICs [59]. Online meeting has also been used for collaborations between geographically distant programs to discuss cases, new initiatives, and research projects [11]. While the use of virtual platforms increased the accessibility and affordability of educational resources for trainees at LMICs, the uptake of new technologies has been slower in LMICs than in HICs and may be limited by variability in internet speed, time zones, and language barrier [12,44]. Poor internet connectivity could result in high frequency of disconnection, poor sound and image quality, and long question-response lag time [83]. One proposed solution to circumvent these barriers was posting videos on free online platforms, such as YouTube, or those that provide live translations [84]. Furthermore, there is now an effort within the global surgery community to increase access to safe and affordable surgical care and to advocate for strengthening the training system in LMICs as a part of pandemic preparedness strategy [85]. Specifically, the Global Neurosurgery Initiatives started in 2018 may be a resource to assist resident training at LMICs, compensating for loss of training during the pandemic [86]. As the impacts of the COVID-19 pandemic on healthcare system continue to evolve, it is imperative to continually evaluate how resident training is affected and explore novel approaches to improve clinical, surgical, and educational experiences as well as prepare for similar future pandemics.

Future recommendations

COVID-19 may be the deadliest viral outbreak the world has experienced in more than a century, but it will not be the last pandemic in our lifetime. In face of these many uncertainties, programs need to reassess their adaptive strategies, create action plans to improve on areas of deficiency, and integrate sustainable initiatives to supplement resident training, which often requires flexibility, innovation, and creativity. The heterogeneity between the included studies signified the importance of adopting a more individualized approach, one that is tailored to the program, the year of residency, and country. Furthermore, the neurosurgical community at large such as the global neurosurgery initiatives should step up and provide openly accessible educational materials during such pandemics to ensure adequate training for the next generation of neurosurgeons around the globe especially trainees from lower income countries such as Sub-Saharan Africa. As the impacts of the COVID-19 pandemic on healthcare systems continue to evolve, it is imperative to continually evaluate how resident training is affected and explore novel approaches to improve clinical, surgical, and educational experiences as well as prepare for similar future pandemics.

Limitations

This systematic review and meta-analysis have several limitations to consider. First, the patient populations enrolled in the selected articles were mainly from the United States and European countries. Insufficient data from other regions, including countries that were hit hard by COVID-19, could prevent our study from drawing a meaningful conclusion. Second, 32.4% of the selected articles involved the use of surveys and may have not been validated due to the time constraint. Biases, such as subjective questions, voluntary responses, and heterogenous respondents and institutions, could have limited the generalizability of our results. Third, multiple pooled estimates had substantial between-study heterogeneity and publication bias. We suspect that, in the setting of COVID-19 pandemic, variation in state legislation, institutional policy, hospital baseline caseloads, the peak of pandemic between countries, the time periods studies were conducted, and the rate of exposure may all contribute to the heterogeneity. Similarly, each neurosurgical residency program around the globe varies widely in terms of workload, caseload, and structure. This may lead to different adaptability to the pandemic and result in variable impact on the training and learning outcome. Furthermore, no study has compared the operative skill and theoretical knowledge of residents between pre-COVID-19 and COVID-19 periods, and thus the true impact of COVID-19 on neurosurgical training remains unknown. Long-term impacts of the COVID-19 pandemic on neurosurgical trainees, including operative competency, mental health, and career advancement, merit further investigation.

Conclusion

The landscape of neurosurgery worldwide has been dramatically transformed by the COVID-19 pandemic, forcing a paradigm shift in both neurosurgical practice and training. Reduction of clinical and surgical exposure has adversely impacted educational provision, which was more pronounced in LMICs. Nevertheless, these unprecedented challenges have rendered an opportunity to rethink the conventional patterns of residency training and identify potentials for collaboration and increase accessibility and affordability for educational resources. It would be of paramount interest to reevaluate, in the near future, the impact of these newly adopted technologies and educational resources on neurosurgical training.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Ethical considerations

This study was not research involving human subjects and that their review and approval was not required.

Data sharing

The data that support the findings of this study are available from the corresponding author on reasonable request.

Potential conflicting interests

All authors have no personal, financial, or institutional interest in the materials or devices described in this manuscript.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Editor: DR B Gyampoh