Predicting the unpredictable: A retrospective cohort study to determine the accuracy of estimated operative duration in orthopedic surgery

Abstract

BACKGROUND

Annually, millions of major surgical procedures are performed worldwide, and precise estimation of operative duration is crucial for surgical planning and operating room (OR) scheduling. Both overestimation and underestimation of procedure time can reduce the efficiency of OR utilization. Therefore, this study investigated the prevalence and factors associated with overestimation and underestimation of the duration of orthopedic surgeries.

AIM

To assess the accuracy of estimated operative duration for orthopedic surgeries by determining the prevalence of overestimation and underestimation and identifying factors associated with these discrepancies.

METHODS

This retrospective, single-center cohort study included patients undergoing orthopedic surgeries at a tertiary hospital. Data were collected from medical records before and after each procedure. Patient demographics, medical status, diagnosis, anesthesia start and end times, estimated and actual surgery durations, and length of hospital stay were analyzed.

RESULTS

Overall, 209 operations were reviewed. The patients’ mean age was 34.8 ± 21.1 years, with 51.2% being male. The mean duration of anesthesia was 3.60 hours, while the mean operating time was 4.37 hours. Underestimation of operative duration by more than 30 minutes was observed in 46.4% of the patients, while the operative time was overestimated for rest of the operations. Anesthesia duration and patient sex were key factors influencing the operative time.

CONCLUSION

In conclusion, overestimation and underestimation of scheduled operation times represent a widespread concern that warrants further attention. Patient comorbidities are the most common factors influencing operation duration, emphasizing the importance of considering patient-specific factors, individualized scheduling, and operation time estimation on a case-by-case basis.

Key Words: Operating room; Operation time; Orthopedic surgeon; Orthopedic surgery; Surgery

Core Tip: Accurate prediction of surgical duration is essential for efficient operating room (OR) scheduling. This study found that both overestimation and underestimation of orthopedic surgery times are common, with surgery times being underestimated for nearly half of all procedures. Patient comorbidities were identified as a key factor influencing operative duration, highlighting the need for individualized, case-based time estimation to improve OR efficiency and surgical planning.

INTRODUCTION

Each year, 234 million major surgical procedures are performed worldwide, emphasizing the fundamental role of surgical care in global health and comprehensive healthcare delivery[1]. Accurate estimation of surgical duration is crucial for surgical planning and an integral factor in operating room (OR) scheduling. Both overestimation and underestimation of procedure time negatively impact the efficiency of OR utilization. Underutilization of the OR and prolonged waiting lists are significant adverse effects of overestimation of surgery time, whereas underestimation can lead to prolonged procedures, delays, and ultimately cancellation of scheduled cases[2]. Moreover, prolonged duration of scheduled surgery results in extended preoperative fasting, adversely affecting patients through dehydration and increased fluid and blood requirements[2-4].

Prolonged surgeries are common across all surgical specialties and increase the risk of postoperative complications[5-8]. In orthopedic procedures, intraoperative delays occur frequently and contribute to a higher incidence of complications[9]. A systematic review concluded that in up to 5.6% of the studied cases, the operative time exceeded the scheduled time by an additional 1.6-1.7 hours[6]. Surgical site infection is the most frequently reported postoperative complication related to prolonged operative time[6]. Other reported complications include increased risks of readmission, reoperation, and delirium[10,11]. In addition, complications can occur among surgeons, particularly shoulder pain[12].

In Saudi Arabia, there is a paucity of research addressing the accuracy of operative time estimation, especially in the field of orthopedic surgery. Understanding the factors that contribute to this problem is crucial for optimizing postoperative outcomes and productively utilizing ORs. This gap in the literature underscores the need for further investigation. Hence, the present study aimed to investigate the prevalence of overestimation and underestimation of operative duration in patients who underwent scheduled orthopedic surgeries and to analyze the associated intraoperative factors. We hypothesized that the preoperative waiting times and estimated durations for orthopedic procedures frequently differ from the actual durations, resulting in significant overestimation or underestimation of the intraoperative duration and higher rates of postoperative complications.

MATERIALS AND METHODS

This retrospective, single-center cohort study included data from patients undergoing orthopedic surgeries in a tertiary hospital over a study period of 12 months. After receiving Institutional Review Board approval from Imam Abdulrahman Bin Faisal University, data were collected from the documented medical records before and after each procedure. Patient demographics, medical status, diagnosis, anesthesia start and end times, estimated (predicted based on the basis of surgeon experience) and actual surgery durations, and length of hospital stay were analyzed. Anesthesia duration was defined as the total time spent by the anesthesia team on induction and preparation for patient recovery, excluding the intraoperative period. The authors classified each procedure duration as underestimated if the actual operating time exceeded the estimated duration by more than 30 minutes and as overestimated if it was less than the estimated duration by 30 minutes or more. This study included all orthopedic procedures performed between March 2022 and March 2023 at this institution. The required sample size was estimated via the Raosoft online sample size calculator (Raosoft Inc., Seattle, WA, United States). Assuming a 95%CI, a 5% margin of error, and a maximum variability assumption (P = 0.5) to provide the most reliable estimate, the minimum recommended sample size was 175 cases. This calculation was based on the number of orthopedic surgeries performed at King Fahad Hospital of the University over the 12-month study period. Ultimately, 209 cases were included in the final analysis, exceeding the calculated sample size and thereby maintaining adequate statistical power to meet the study objectives. Patients who did not have medical records or who underwent surgeries not performed by orthopedic surgeons were excluded. Finally, data for a total of 209 orthopedic procedures, which met the inclusion criteria, were collected from the electronic documentation of patients’ medical reports. The data were analyzed using the software program Statistical Package for the Social Sciences version 26 (Armonk, NY, United States). Descriptive statistics are presented as n (%) for all categorical variables, while the median (interquartile range) was used to describe continuous variables. Factors affecting the precision of surgical duration prediction were determined via the χ² tests and independent sample t tests. Significant results were then tested through multivariate logistic regression analysis to identify the independent predictors of overestimation and underestimation of scheduled surgery duration, with corresponding odds ratios and 95%CIs. A P value < 0.05 was considered significant.

RESULTS

Overall, 209 patients were included in this study. Table 1 presents the patient characteristics. The patients’ mean age was 34.8 ± 21.1 years, and 51.2% were male. Elective surgeries were the most common type of surgery (89.5%). Nearly half (48.8%) of the patients had no medical comorbidities. Surgical duration underestimation by at least 30 minutes was observed in 46.4% of the patients. The mean duration of anesthesia was 3.60 hours, while the mean intraoperative duration was 4.37 hours. The mean length of hospitalization was 2.59 days. As shown in Figure 1, the most common subspecialty was arthroplasty (20.1%), followed by pediatrics (17.2%) and trauma (16.3%).

Figure 1 Surgical procedures performed during the study period stratified by subspecialties.

Table 1 Patient characteristics, n (%)/mean ± SD.

Study data
Age (years)	34.8 ± 21.1
Gender
Male	107 (51.2)
Female	102 (48.8)
Method of surgery
Elective	187 (89.5)
Emergency	22 (10.5)
Patient medical status
Not known to have medical illness	102 (48.8)
HTN	6 (2.9)
DM	5 (2.4)
Multimorbidity	31 (14.8)
Others	65 (31.1)
Precision of surgical duration
Underestimated duration (> 30 minutes)	97 (46.4)
Overestimated duration (≤ 30 minutes)	112 (53.6)
Anesthesia duration in hours	3.60 ± 1.86
Intraoperative duration in hours	4.37 ± 2.76
Length of hospital stay in days	2.59 ± 1.64

HTN: Hypertension; DM: Diabetes mellitus.

As presented in Table 2, underestimated surgical procedures were significantly more common in male patients (P = 0.021) and patients with comorbidities (P = 0.010). No significant associations were detected between the precision of predicted surgical duration and age, subspecialty, method of surgery, or length of hospital stay (P > 0.05).

Table 2 Factors influencing the accuracy of estimated surgical duration, n (%)/mean ± SD.

Factor	Precision of surgical duration		P value
	Underestimated (n = 97)	Overestimated (n = 112)
Age (years)	33.9 ± 20.4	35.6 ± 21.8	0.571
Gender
Male	58 (59.8)	49 (43.8)	0.021
Female	39 (40.2)	63 (56.3)
Subspecialty
Pediatrics	14 (14.4))	22 (19.6)	0.700
Pediatrics trauma	3 (3.1)	2 (1.8)
Trauma	19 (19.6)	15 (13.4)
Upper extremity	08 (08.2)	14 (12.5)
Sport	15 (15.5)	12 (10.7)
Arthroplasty	19 (19.6)	23 (20.5)
Foot and ankle	11 (11.3)	16 (14.3)
Spine	8 (8.2)	8 (7.1)
Method of surgery
Elective	85 (87.6)	102 (91.1)	0.419
Emergency	12 (12.4)	10 (8.9)
Patient medical status

Multivariate regression analysis (Table 3) predicted a lower risk of underestimation of surgical duration in female patients by at least 71% compared with that in male patients (adjusted odds ratio = 0.294; 95%CI: 0.128-0.676; P = 0.004). However, for every point increase in anesthesia duration, the risk of surgical duration underestimation increased by at least 2.7 times (adjusted odds ratio = 2.682; 95%CI: 2.035-3.535; P < 0.001). However, the patients’ medical status and intraoperative duration had no significant effect on the underestimation of surgical duration after adjustment in a regression model (P > 0.05).

Table 3 Independent factors influencing the accuracy of estimated surgical duration.

Factor	AOR	95%CI	P value
Gender
Male	Reference
Female	0.294	0.128-0.676	0.004
Patient medical status
Not known to have medical illness	Reference
Comorbidities	0.522	0.235-1.159	0.110
Anesthesia duration in hours	2.682	2.035-3.535	< 0.001
Intraoperative duration in hours	1.175	0.930-1.486	0.177

AOR: Adjusted odds ratio.

DISCUSSION

Operative timing and prolonged scheduled operations are global dilemmas. Intraoperative delays affect not only patients but also healthcare workers and the efficiency of resource utilization[13]. With the continuous rise in global surgical volume over recent decades, the precision of operative time estimation has become increasingly vital. The operative duration itself is an independent yet modifiable risk factor for postoperative complications[1]. Visser et al[14] identified underestimation of the operative time as one of the three major predictors of surgical complications. Similarly, a recent systematic review and meta-analysis concluded that prolonged operation time is significantly associated with higher complication rates, noting that every 30-minute delay increases the likelihood of adverse outcomes by approximately 14%[6]. Falzetti et al[7] further demonstrated that delays in surgical duration contribute directly to postoperative complications and, ultimately, longer hospital stays. Furthermore, Butler et al[15] measured and evaluated the accuracy of surgical time estimation by orthopedic surgeons, anesthesiologists, staff, and nurses and found that orthopedic surgeons made the most accurate estimations for surgery duration. This finding supports and encourages orthopedic surgeons to estimate and input the surgery duration estimations to help achieve more effective operative scheduling for the benefit of patients and the OR staff.

In the present study, the prevalence of underestimated duration for scheduled orthopedic operations was 46.4% among the 209 cases analyzed. Nearly half of the operations exceeded their scheduled times by more than 30 minutes. These findings are consistent with a study conducted by the Department of Orthopedic Surgery at the Nagoya University Graduate School of Medicine in Japan, which examined multiple spinal procedures and reported that 49% (n = 488) of the procedures required more time than scheduled, with a delay of more than 2 hours occurring in some cases[16]. Similarly, a study in Saudi Arabia encompassing all surgical specialties reported that 54.6% of orthopedic surgeries surpassed the scheduled duration[17]. Furthermore, recent evidence suggests that the use of magnesium sulfate as an adjuvant in anesthesia for postoperative pain management is associated with a prolonged total operative time[18].

Among the subspecialties, arthroplasty was the most frequently performed procedure during the data collection period, accounting for 42 surgeries (20.1%), followed by pediatric orthopedic surgeries with 36 procedures (17.2%). Pediatric orthopedic trauma and spine operations were the least common operations, accounting for 5 and 16 operations, respectively. However, underestimation of operative time was most frequently observed among pediatric orthopedic trauma patients. Underestimation of operative time by more than 30 minutes was also identified in nearly half of the adult trauma and sports surgeries. In contrast, upper extremity surgeries demonstrated the most accurate operative time predictions among all subspecialties followed by foot and ankle surgeries. Overall, upper extremity and foot and ankle surgeries were performed within the scheduled time without intraoperative delay. Although variations in operative duration were observed among orthopedic subspecialties, they did not reach statistical significance (P = 0.700). To date, no prior research has reported comparable data to allow direct comparison.

In addition, our analysis revealed that underestimation of operative duration was significantly more common among male patients (P = 0.021) and patients with comorbidities (P = 0.010). This has been attributed to the variation in tissue composition, inflammation, healing process, and immune response in male patients[19]. Other studies have also shown that male patients are more likely to have delayed presentation and higher baseline risk, because they have increased prevalences of smoking, alcohol use, and cardiovascular risk[20]. Additionally, extended operations in patients with comorbidities demonstrate that patient characteristics can significantly affect the operating time. These results are in line with those of Rohrer et al[10] who concluded that comorbidities are associated with a greater risk of prolonged surgery duration and are an independent risk factor for readmission and reoperation. This finding is supported by a previous study conducted among 454 patients who underwent unilateral primary total knee arthroplasty in 2012 in the United States. In that study, procedures for class III obese patients required significantly more operating time than did procedures for patients with a normal body mass index (BMI)[21]. In contrast, in 2015, a Saudi study that analyzed 204 knee replacement cases concluded that BMI and other patient comorbidities, including diabetes mellitus, hypertension, and dyslipidemia, were not associated with operating time[22]. Based on the results of the present study, we still suggest that patient comorbidities significantly affect operative time. Thus, considering the unique factors of each case and calculating the estimated operating time at an individual level can facilitate more accurate scheduling, effectively enhancing OR usage and reducing case cancellations.

Despite our efforts to explore this multifaceted issue, we must acknowledge several limitations of the present study. A notable limitation is the lack of control for potential confounding variables. While our analysis accounted for factors such as patient sex, comorbidities, and anesthesia duration, other important confounders such as surgical complexity, surgeon experience, patient BMI, and American Society of Anesthesiologists classification were not evaluated because of constraints in the retrospective study design. Intraoperative logistical delays, including the unavailability of surgical instruments or delays related to radiological coordination, were not captured in our data. These unmeasured factors may independently influence the operative duration and contribute significantly to intraoperative delays. Emergency procedures, particularly those for trauma cases, may involve additional variability due to associated injuries, unanticipated intraoperative findings, and urgent decision-making, which can influence the accuracy of operative time estimation. Similarly, factors such as BMI, surgeon experience, procedural complexity, and intraoperative adverse events may contribute to time variation. Because these variables were not consistently available in our retrospective dataset, they could not be analyzed in the present study. Future prospective studies should evaluate these factors in detail and compare elective and emergency procedures separately to better define modifiable causes of operative time misestimation. Future prospective research should aim to comprehensively incorporate these clinical and procedural variables, enabling better control for confounding factors and enhancing the generalizability of findings. This paper provides valuable insights into the overlooked challenges affecting the efficient utilization of ORs.

CONCLUSION

Underestimation of the time required for scheduled operations is a universal concern that necessitates further attention. This issue significantly impacts both patients and the healthcare system. Our findings indicate that anesthesia duration and patient sex are key factors influencing operative time. These findings highlight the value of integrating anesthesia-related factors and patient sex into surgical planning for orthopedic procedures. The implementation of predictive models or computerized scheduling systems that incorporate these variables can aid the accuracy of surgery time estimation. Improved accuracy will ultimately minimize procedure delays and cancellations, thereby improving patient safety and satisfaction. Further research is recommended to evaluate the postoperative complications associated with operative duration and refine strategies to enhance surgical care. In addition, future studies should focus on the impact of logistical factors, such as the availability of surgical instruments and radiological coordination, which may play a significant role in the accuracy of the estimated surgical duration.