Impact of surgical start time on error rates in cataract surgery: A retrospective cohort study

Abstract

BACKGROUND

Surgical errors remain an ongoing concern in ophthalmology, particularly in high-volume procedures such as cataract surgery, where even low complication rates translate into significant patient impact. Factors such as surgical start time and surgeon fatigue have been proposed as potential contributors to intraoperative complications, though evidence within ophthalmology remains limited and inconsistent. While some studies in other surgical specialties suggest time-of-day effects on outcomes, data specific to ophthalmology are sparse, and the extent to which surgical timing influences performance remains unclear.

AIM

To assess whether the timing of surgery influences the rate of surgical errors and complications during cataract operations.

METHODS

A retrospective cohort study was conducted at a university center in Sherbrooke, Québec, Canada. We included adults who underwent primary cataract surgery from 2006-2023. Electronic medical records were reviewed for baseline demographics, intraoperative parameters, and postoperative outcomes. Complication rates were compared across hourly time intervals. Continuous and categorical variables were analyzed using appropriate descriptive statistics, and associations between surgical timing and complication rates were assessed using a modified Poisson regression model with statistical significance set at P < 0.05.

RESULTS

A total of 32762 cataract surgeries were included in the analysis, with an overall complication rate of 1.14%. Temporal analysis across the surgical day (7:00 to 15:00) demonstrated no statistically significant variation in complication rates, which ranged from 0.8% to 1.7%. The highest complication rate was observed at 13:00 (1.66%), while the lowest occurred at 15:00 (0.76%); however, these differences were not statistically significant. No consistent trend in complication rates was observed as the day progressed, and confidence intervals overlapped across all time periods.

CONCLUSION

Surgical start time does not appear to significantly influence intraoperative complication rates in cataract surgery. These findings support the safety of current scheduling practices within standard operating hours and suggest that concerns regarding time-of-day effects may be overstated in this context. Future research should focus on identifying other modifiable factors that may contribute to surgical errors and optimizing strategies to further enhance patient safety and surgical outcomes.

Key Words: Cataract surgery; Surgical start time; Intraoperative complications; Surgical errors; Surgeon fatigue; Ophthalmology; Retrospective; Patient safety; Phacoemulsification; Surgical scheduling

Core Tip: This large retrospective cohort study evaluates whether surgical start time influences intraoperative complication rates in cataract surgery. Analyzing over 32000 cases, we found no significant variation in complication rates across the surgical day. These findings challenge assumptions that surgeon fatigue or time of day meaningfully impact surgical performance within standard operating hours. By isolating surgeon-related factors in a high-volume, standardized procedure, this study provides robust evidence supporting current scheduling practices and highlights the need to investigate alternative contributors to surgical error.

INTRODUCTION

Despite remarkable progress in the field of ophthalmic surgery, the potential for errors and complications remains an ongoing concern[1]. Factors such as surgical start time and surgeon fatigue could potentially impact the rate of errors[2,3].

An extensive review of the literature revealed only three prior studies in ophthalmology examining the influence of surgical timing on outcomes[4-6]. Gerstmeyer and Lehrl[4] reported a possible mid-morning increase in surgical errors, which they attributed to physiological fatigue. Gupta and Taravati[5] later analyzed over a thousand cataract cases and found no statistically significant association between case order and complication rates, though a slight trend towards increased complications later in the day was observed. Shields et al[6] investigated the effect of surgical timing and trainee involvement in pars plana vitrectomy. They reported no meaningful difference in postoperative adverse events across morning, afternoon, or evening time slots. Though the risk of postoperative vitreous hemorrhage was significantly higher in afternoon and evening cases. The presence of a surgical fellow did not influence their complication rates or operative time.

Several additional studies have explored the relationship between surgical start time and complications within other surgical specialties, including general surgery, neurosurgery[7-9], gynecologic surgery[10-12], and spine surgery[13-15]. These studies predominantly focus on the association between surgical start time and increased complications. They do not specifically examine surgeon errors, as complications can be influenced by various factors such as the performance of the entire surgical team, including nurses and anesthetists, and patient-related factors. Moreover, the findings from these studies are mixed, with some suggesting a significant effect of surgical start time on complications, while others reporting no effect.

Given the scarcity of robust data in ophthalmology and the potential impact of surgical start time and surgeon fatigue on patient outcomes, it is imperative to further investigate this relationship. Cataract surgery, distinguishable for its high frequency, consistent distribution, short duration and motor and cognitive demands represents an ideal procedure for evaluating the impact of surgeon-related errors. Moreover, reduced reliance on other team members, such as the absence of anesthetists, further enhances its suitability for this purpose. This study aims to fill the gap in the literature and provide evidence-based insights for ophthalmology surgery practices. By investigating the relationship between surgical start time and error rates, this study aims to enhance our understanding of the factors contributing to errors in cataract surgery and ultimately aid in the mitigation of surgical errors in all surgical specialties. The primary objective of this study was to identify a relationship between surgical start time and intraoperative error/complication rates in cataract surgery. Secondary objectives were to identify additional variables that may be associated with errors [e.g., patient’s body mass index (BMI), presence of trainee, day of the week, etc.], and to offer evidence-based recommendations for surgical scheduling to minimize errors in ophthalmology surgery and raise awareness of surgeon’s fatigue. We hypothesized that later surgical start times would be significantly associated with intraoperative errors and/or complication rates in cataract surgery within standard operating hours.

MATERIALS AND METHODS

Study design and patient selection

A retrospective cohort study was conducted using electronic medical records in our university center at the Université de Sherbrooke, Québec, Canada. Patients aged 18 years and older who underwent primary cataract surgery between 2006 and 2023 were included in the study. Those with traumatic cataracts as well as those with ocular comorbidities complexifying the surgical procedure were excluded.

In the context of our study, we focused exclusively on direct surgical complications related to cataract surgery, specifically those attributable to intraoperative surgeon error. This approach deliberately omits indirect complications, such as post-operative hyphema, corneal edema, or ocular hypertension. This decision was based on the potential variability in how these conditions are documented, which could affect the consistency of data.

Data collection

A total of 32762 cataract surgeries done in 24834 patients were included in the study. All electronic medical records were thoroughly analyzed. Patient demographics including but not limited to age, sex, BMI, and the Charleston Comorbidity Index were noted. Pre-operative measures included ocular comorbidities, best-corrected visual acuity, intra-ocular pressure and lens grade status. Intra-operative measures included the type of surgery (i.e., unilateral or bilateral), the eye treated, the type of intraocular lens implanted, the presence of a trainee, the effective phacoemulsification time and the date of the surgery. Intra-operative complications as well as the treatment and efficacy were noted. Post-operative evaluation including the best-corrected visual acuity and intra-ocular pressure were recorded.

Outcome measures

The primary outcome measure was the rate of errors and complications per surgical start time slot (e.g., 8:00-9:00, 9:00-10:00, 10:00-11:00, etc.). The secondary outcome measures consisted of additional factors which may influence complication rates (e.g., age, BMI, etc.)

Statistical analysis

All data were analyzed using descriptive and inferential statistical methods. Continuous variables (e.g., age, BMI) were summarized as medians with interquartile ranges, and categorical variables as n (%). Comparisons between surgeries with and without complications were assessed using a modified Poisson regression model to estimate relative risks, with binary intraoperative complication status (occurrence of a complication vs non-occurrence of a complication) as the dependent variable and baseline demographic/clinical characteristics as independent variables. A two-sided P < 0.05 was considered statistically significant. This approach was utilized because it is appropriate for binary outcomes in retrospective cohort data and provides stable estimation in large datasets.

Complication rates were analyzed by hourly surgical start times (7:00-15:00) and displayed with 95% confidence intervals. Variation across time periods were evaluated using the same model. The large sample size (n = 32762) ensured sufficient power.

RESULTS

Within our extensive cohort of 32762 cataract surgeries (n = 32762), we observed a total of 373 intraoperative direct complications, yielding an overall complication rate of approximately 1.14%. As shown in Table 1, the top three complications were posterior capsule ruptures, zonular breaks as well as anterior capsule ruptures. As shown in Table 2, baseline characteristics between both groups were statistically insignificant.

Table 1 Frequency of complication rates.

Complication	Frequency, n (%)
Posterior capsule rupture	281 (75.3)
Zonulysis	44 (11.8)
Anterior capsule rupture	34 (9.1)
Corneal injury	8 (2.1)
Intra-ocular IOL break	4 (1.1)
Incorrect IOL	1 (0.3)
Iridodialysis	1 (0.3)

IOL: Intraocular lens.

Table 2 Preoperative baseline demographics and clinical characteristics, median (interquartile rage).

	Surgeries without complications (n = 32389)	Surgeries with complications (n = 373)	P value
Median age	73 (67-79)	73 (66-80)	0.811
Sex (female, male)	18713, 13669	210, 163	0.563
BMI	26.2 (24-31)	28 (24-31)	0.761

P < 0.05 is considered statistically significant. P-value was calculated using the modified Poisson model. BMI: Body mass index.

Primary outcomes

In examining the correlation between surgical start times and the incidence of complications in cataract surgery, we present our findings in Figure 1. This bar chart delineates the proportion of surgeries with complications per hour, starting from 7:00 to 15:00. Each bar signifies the percentage of surgeries that resulted in complications during the respective time slot (e.g., 7:00 represents the period from 7:00 to 8:00), accompanied by the absolute numbers indicating the count of complicated cases juxtaposed with the total number of surgeries conducted in that hour.

Figure 1  Complications rates in cataract surgery by surgical start time period.

A pattern of fluctuation in complication rates across the different time slots is observed, without a clear directional trend as the day progresses. The complications rates remain between 0.8% and 1.7%. The analysis revealed that the highest complication rate was recorded at 13:00 (1.66%), while the lowest rate was observed at 15:00 (0.76%). The presence of error bars spanning each bar in the figure reflects the confidence intervals for each period's complication rate, suggesting that the observed variations in complication rates between different surgical start times do not reach statistical significance. The sample size notably diminishes as the day unfolds, with a significant decrease from 8697 surgeries at the beginning of the operational hours to 528 by late afternoon.

Secondary outcomes

The secondary objective of the present study was to conduct a comparison of patient demographics in order to identify any additional factors that may influence complication rates.

As presented in Table 2, all demographic values were found to be statistically insignificant, indicating that there were no meaningful differences across the age, BMI and sex of the patients examined. This lack of statistical significance suggests that these demographic factors do not have a substantial impact on complication rates.

DISCUSSION

In this large retrospective cohort of over 32000 cataract surgeries, surgical start time was not significantly associated with intraoperative complication rates. Though complication rates varied modestly across the hourly time slots, no consistent temporal trend was observed, suggesting that cataract surgery performed within standard operating hours does not carry increased risk based on time of day alone.

The study's robust sample size, sets a precedent as one of the most extensive in recent years, particularly given the relatively low incidence rate of complications in cataract surgery, which typically falls around 1%-2%[16,17]. Such a substantial cohort is pivotal, providing the statistical power necessary to circumvent the risk of a type II (beta) error. This statistical robustness allows us to state with confidence that the absence of a statistically significant difference in complication rates across various surgical start times is not a consequence of inadequate sample size or power. Rather, it likely reflects an authentic lack of temporal effect on the incidence of complications.

Our comprehensive data analysis enhances this study further. By integrating patients’ backgrounds and preoperative characteristics, we have bolstered the reliability of our findings. This thorough approach ensures that any differences observed - or lack thereof - are attributable to our primary variable, surgical start time, rather than confounding or secondary factors.

Our findings align with Gupta and Taravati[5] and Shields et al[6], who likewise reported no significant association between surgical timing and complication rates, though Shields et al[6] did observe a significantly higher risk of postoperative vitreous hemorrhage in afternoon/evening cases.

While the strengths of our study are reinforced by the considerable sample size and the comprehensive data analysis, it is necessary to acknowledge the constraints inherent in our methodology. The retrospective nature of our study design, reliant on the analysis of electronic medical records, presents certain inherent limitations. The accuracy and completeness of retrospective data can introduce biases that might compromise the validity of our results. Additionally, the inherent constraints of a retrospective framework preclude us from establishing a causal relationship between surgical start times, surgeon fatigue, and error rates definitively.

Confounding factors also pose a significant limitation. Despite our efforts to adjust for variables such as patient demographics and pre-operative characteristics, the potential for residual confounding cannot be entirely excluded. This might affect the perceived relationship between surgical start times and error rates.

While our study did not examine potential variation by day of the week or season, prior work by Kork et al[13] across multiple surgical disciplines found no significant differences in postoperative mortality across these factors in a large mixed surgical cohort. Such findings suggest that broader temporal variables (such as weekday or seasonal effects) may likewise have minimal influence on surgical outcomes.

Another concern is the possibility of selection bias, stemming from our exclusion criteria. By excluding patients with prior ocular surgeries, trauma, or comorbidities that could affect surgical outcomes, we risk creating a cohort that may not fully represent the broader demographic of patients undergoing cataract surgery. Furthermore, by focusing exclusively on patients operated on in hospital operating rooms, we potentially limit the external validity of our study, as we do not account for surgeries performed in outpatient settings.

Cataract surgery is a unique procedure that enables us to evaluate the relationship between surgical start time, fatigue, and error rates due to its ability to isolate surgeon factors from other variables. Its high frequency, consistent distribution, short duration, high motor and cognitive demands, and lesser dependence on other team members’ performance make it an ideal setting for this investigation.

The implications of our study extend beyond ophthalmology to other surgical specialties, as the findings can provide valuable insights into the general impact of surgical start time and other factors on error rates. By identifying the factors contributing to errors in surgery, we can inform future research and interventions aimed at reducing error rates across various surgical fields.

CONCLUSION

The present study sought to elucidate the impact of surgical start times on complication rates in cataract surgery, an area previously underexplored in ophthalmic research. Over a robust sample of 32762 surgeries, our investigation revealed an overall complication rate of 1.14%. Crucially, the temporal analysis across the span of the typical surgical day, from 7:00 to 15:00, did not demonstrate a statistically significant variation in complication rates, which remained between 0.8% and 1.7%. The highest recorded complication rate occurred at the 13:00 hour period, with a rate of 1.66%, while the lowest was at 15:00, with a rate of 0.93%.

These findings suggest that, within the observed time frame, the start time of cataract surgeries does not significantly affect the likelihood of complications. The absence of a significant temporal trend in complication rates suggests that the start time of cataract surgeries, within the morning and afternoon hours (between 7:00 and 16:00), does not substantially influence the likelihood of complications. This finding does not support the hypothesis that surgeon fatigue accumulates in a manner that affects surgical precision and safety, at least within the time range studied. As such, our results provide empirical support for the current scheduling practices in cataract surgery, which do not appear to elevate the risk of complications based on the time of day the surgery is performed.

The implications of these findings are twofold. Firstly, they provide reassurance regarding the safety of current cataract surgery scheduling. Secondly, they redirect the focus towards other potential factors that may influence surgical outcomes. As we continue to explore secondary objectives, such as the impact of patient characteristics (e.g., patient’s BMI) or the presence of trainees, we aim to build on our understanding of what contributes to surgical errors.

This study contributes valuable knowledge to the field of ophthalmology and other surgical specialties, suggesting that the timing of surgeries need not be a primary concern for surgical planning with regard to complication rates, as long as it falls into normal working hours. It also lays the groundwork for future research to explore other dimensions of surgical performance and patient safety. By continuing to examine the factors that contribute to complications, we uphold our commitment to improving surgical outcomes and the well-being of our patients.