CUSUM learning curves: what they can and can't tell us

Evaluation of learning curves for contact laser vaporization of the prostate using the 980 nm diode laser for benign prostatic hyperplasia

OBJECTIVES

We investigated the background of patients who underwent contact laser vaporization of the prostate (CVP) surgery and the learning curve of the operators.

METHODS

A total of 207 patients who underwent CVP surgery for benign prostatic hyperplasia between August 2018 and March 2023 were included in this study. Patient background, perioperative results, pre- and postoperative urinary flow tests, and complications were collected retrospectively.

RESULTS

We enrolled 12 doctors who were divided into expert (five doctors) and novice (seven doctors) groups based on the number of TURP experiences before CVP. The median patient age was 73 years (51-92 years) and prostate volume was 56 cc (15-190 cc) with no difference between the expert and novice groups. Complications included urinary retention (eight cases), hematuria (four), urinary tract infection (four), intraoperative perforation (two), and postoperative stricture (one). Both cases of intraoperative perforation occurred in the novice group. The expert group had a significantly shorter operative time (38 vs. 66 min) and a higher operative efficacy of prostate volume divided by operative time (1.43 vs. 0.88 cc/min). Postoperatively, IPSS, quality of life scores, and postvoid residual urine volume decreased, and maximal flow rate increased; however, there was no significant difference between the groups. The expert group showed stable operative time and operative efficacy after about five to eight cases, while the novice group showed stable after about 15 cases.

CONCLUSIONS

Our findings suggest that CVP was safely performed at our hospital, and operators with limited experience in TURP can achieve stable perioperative results.

Cumulative sum (CUSUM) analysis: A 10 year experience of outcome monitoring in a regional burn centre

INTRODUCTION

Outcome monitoring can identify fluctuations in healthcare. Cumulative sum (CUSUM) analysis can detect when mortality deviates from expected, allowing early intervention through targeted audit. We present a 10-year experience of the prospective use of CUSUM methodology in a regional burn centre.

METHOD

Prospective outcome monitoring was conducted for all admissions to the intensive care unit between 2012 and 2022. The revised Baux score was used for mortality risk prediction. Risk-adjusted CUSUM charts tracked mortality against that predicted by the revised Baux score. Deviations from expected outcomes triggered detailed structured analysis. Learning outcomes were identified from internal and external governance groups.

RESULTS

CUSUM analysis was triggered on eight occasions: one paediatric (excess deaths), six adult (four excess survivors, two excess deaths) and one elderly (excess survivors). Detailed analysis identified areas for continuous improvement and positive themes from excess survivors.

CONCLUSION

The use of CUSUM as an early warning trigger stimulates assessment of practice and critical appraisal of factors contributing to unexpected mortality or survival. The revised Baux score at its foundation needs to be carefully considered but remains a valid model. One benefit is positive reinforcement of team cohesion and morale during periods of care excellence leading to excess survivors.

Charting Proficiency: The Learning Curve in Robotic Hysterectomy for Large Uteri Exceeding 1000 g

Background/Objectives: This study evaluates the safety and surgical outcomes of performing robotic hysterectomy on uteri weighing over 1000 g, with a focus on the surgeon's learning curve. Methods: A retrospective analysis was conducted on 44 patients who underwent hysterectomy by a single surgeon from January 2020 to February 2024 using the DaVinci Xi System. Surgical procedures included total hysterectomy with bilateral salpingectomy, and specimens were removed via transvaginal manual morcellation. Operative times were segmented into docking, console, morcellation, and conversion times. Results: Results indicated an inflection point in the 20th case, suggesting proficiency after 20 surgeries. Comparison between early (Group A, cases 1-20) and later cases (Group B, cases 21-44) showed significant reductions in console time (CT) and morcellation time (MT) in Group B, leading to a shorter overall operative time (OT). Although estimated blood loss was higher in Group A, it was not statistically significant. Hemoglobin differences were significantly higher in Group B. No significant differences were observed in transfusion rates, postoperative analgesic usage, or complications between the groups. Conclusions: The study concludes that robotic hysterectomy for large uteri is safe and that surgical proficiency improves significantly after 20 cases, enhancing overall outcomes.

Learning curve of laparoscopic appendectomy in a low-resource setting: a cumulative sum analysis of operative length

BACKGROUND

Cumulative sum (CUSUM) analysis is a valuable tool for quantifying the learning curve of surgical teams by detecting significant changes in operative length. However, there is limited research evaluating the learning curve of laparoscopic techniques in low-resource settings. The objective of this study is to evaluate the learning curve for laparoscopic appendectomy within a single surgical team in Senegal.

METHODS

This was a single-center prospective study conducted from May 1, 2018, to August 31, 2023 of patients who underwent laparoscopic appendectomy at a tertiary care institution in West Africa. The AAST classification was used to describe the severity of appendicitis. Parameters studied included age, sex, operative length, conversion rate, and postoperative outcomes. To quantify the learning curve, CUSUM analysis of operative length was performed.

RESULTS

A total of 81 patients were included. The mean age was 26.7 years (range 11-70 years) with a sex ratio of 1.9. Pre-operative severity according to AAST was Grade I in 75.4% (n = 61), Grade III in 7.4% (n = 6), Grade IV in 6.1% (n = 5), and Grade V in 11.1% (n = 9). Conversion occurred in 5 cases (6.1%). The average operative length was 76.8 min (range 30-180 min) and the average length of hospitalization was 2.7 days (range 1-13 days). Morbidity was observed in 3.7% (n = 3) and there were no deaths. The CUSUM analysis showed that a steady operative length was achieved after 28 procedures, with decreasing operative lengths thereafter.

CONCLUSION

Surgeons in our setting overcame the learning curve for laparoscopic appendectomy after performing 28 procedures. Moreover, laparoscopic appendectomy is safe and feasible throughout the learning curve. CUSUM analysis should be applied to other laparoscopic procedures and individualized by surgical teams to improve surgical performance and patient outcomes in low-resource settings.

Robot-Assisted Radical Prostatectomy (RARP) Trifecta Learning Curve for Surgeons with Previous Experience in Laparoscopy

Background and Objectives: Robot-assisted radical prostatectomy (RARP) is a complex surgery with a steep learning curve (LC). No clear evidence exists for how previous laparoscopic experience affects the RARP LC. We report the LC of three surgeons with vast experience in laparoscopy (more than 400 procedures), analyzing the results of functional and oncological outcomes under the "Trifecta" concept (defined as the achievement of continence, potency, and oncological control free of biochemical recurrence). Materials and Methods: The surgical experience of the three surgeons from September 2021 to December 2022, involving 146 RARP consecutive patients in a single institution center, was evaluated prospectively. Erectile disfunction patients were excluded. ANOVA and chi-square test were used to compare the distribution of variables between the three surgeons. LC analysis was performed using the cumulative sum control chart (CUSUM) technique to achieve trifecta. Results: The median age was 65.42 (±7.34); the clinical stage were T1c (68%) and T2a (32%); the biopsy grades were ISUP 1 (15.9%), ISUP 2 (47.98), and ≥ISUP 3 (35%). The median surgical time was 132.8 (±32.8), and the mean intraoperative bleeding was 186 cc (±115). Complications included the following: Clavien-Dindo I 8/146 (5.47%); II 9/146 (6.16%); and III 3/146 (2.05%). Positive margins were reported in 44/146 (30.13%). The PSA of 145/146 patients (99%) at 6 months was below 0.08. Early continence was achieved in 101/146 (69.17%), 6-month continence 126/146 (86%), early potency 51/146 (34.9%), and 6-month potency 65/146 (44%). Surgeons "a", "b", and "c" performed 50, 47, and 49 cases, respectively. After CUSUM analysis, the "Trifecta" LC peak was achieved at case 19 in surgeon "a", 21 in surgeon "b", and 20 in surgeon "c". Conclusions: RARP LC to accomplish "Trifecta" can be significantly reduced in surgeons with previous experience in laparoscopy and be achieved at around 20 cases.

Improving assessment of procedural skills in health sciences education: a validation study of a rubrics system in neurophysiotherapy

BACKGROUND

The development of procedural skills is essential in health sciences education. Rubrics can be useful for learning and assessing these skills. To this end, a set of rubrics were developed in case of neurophysiotherapy maneuvers for undergraduates. Although students found the rubrics to be valid and useful in previous courses, the analysis of the practical exam results showed the need to change them in order to improve their validity and reliability, especially when used for summative purposes. After reviewing the rubrics, this paper analyzes their validity and reliability for promoting the learning of neurophysiotherapy maneuvers and assessing the acquisition of the procedural skills they involve.

METHODS

In this cross-sectional and psychometric study, six experts and 142 undergraduate students of a neurophysiotherapy subject from a Spanish university participated. The rubrics' validity (content and structural) and reliability (inter-rater and internal consistency) were analyzed. The students' scores in the subject practical exam derived from the application of the rubrics, as well as the rubrics' criteria difficulty and discrimination indices were also determined.

RESULTS

The rubrics´ content validity was found to be adequate (Content Validity Index > 0.90). These showed a unidimensional structure, and an acceptable internal consistency (α = 0.71) and inter-rater reliability (Fleiss' ƙ=0.44, ICC = 0.94). The scores of the subject practical exam practically covered the entire range of possible theoretical scores, showing all the criterion medium-low to medium difficulty indices - except for the one related to the physical therapist position-. All the criterion exhibited adequate discrimination indices (rpbis > 0.39), as did the rubric as a whole (Ferguson's δ = 0.86). Students highlighted the rubrics´ usefulness for learning the maneuvers, as well as their validity and reliability for formative and summative assessment.

CONCLUSIONS

The changed rubrics constitute a valid and reliable instrument for evaluating the execution quality of neurophysiotherapy maneuvers from a summative evaluation viewpoint. This study facilitates the development of rubrics aimed at promoting different practical skills in health-science education.

Design, Fabrication, and Preliminary Validation of Patient-Specific Spine Section Phantoms for Use in Training Spine Surgeons Outside the Operating Room/Theatre

Pedicle screw fixation (PSF) demands rigorous training to mitigate the risk of severe neurovascular complications arising from screw misplacement. This paper introduces a patient-specific phantom designed for PSF training, extending a portion of the learning process beyond the confines of the surgical room. Six phantoms of the thoracolumbar region were fabricated from radiological datasets, combining 3D printing and casting techniques. The phantoms were employed in three training sessions by a fifth-year resident who performed full training on all six phantoms; he/she placed a total of 57 pedicle screws. Analysis of the learning curve, focusing on time per screw and positioning accuracy, revealed attainment of an asymptotic performance level (around 3 min per screw) after 40 screws. The phantom's efficacy was evaluated by three experts and six residents, each inserting a minimum of four screws. Initial assessments confirmed face, content, and construct validity, affirming the patient-specific phantoms as a valuable training resource. These proposed phantoms exhibit great promise as an essential tool in surgical training as they exhibited a demonstrable learning effect on the PSF technique. This study lays the foundation for further exploration and underscores the potential impact of these patient-specific phantoms on the future of spinal surgical education.