Lumbar discectomy: should patient body mass index and abdominal girth be considered when selecting cases suitable for training opportunities? A prospective observational study

Study Design: Prospective observational study. Objectives: To determine whether body mass index (BMI) and abdominal girth are reliable predictors of wound depth and operative duration at lumbar discectomy, to compare the duration of traineeand consultant-performed cases, and to assess the role of patient body habitus when selecting cases for training opportunities. Methods: Pre-operative height and weight, abdominal girth in knee-chest position, and intra-operative wound measurements were taken. Operative duration was recorded using theatre management software. Patients undergoing single-level lumbar discectomy were selected, n=29, M:F ratio 18:11. Mean age was 47.2 years (range 18-83). Discectomy level: L2/3 (1); L3/4 (1), L4/5 (18); L5/S1 (9). Results: Mean BMI was 30.4kg/m (19.7-47.2), abdominal girth 106cm (48-154), wound length 5.4cm (4.56.5), subcutaneous fat depth 1.9cm (0.5-5.2), muscle depth 3.2cm (2.0-4.2), total wound depth 5.0cm (3.59.0), operation duration 47 minutes (24-82). Fifteen patients were clinically obese (BMI >30 kg/m). Fifteen cases were trainee-performed and fourteen consultant-performed. Mean trainee operative duration was 50 minutes (24-77). Mean consultant operative duration was 45 minutes (24-82). Of the cases with BMI >30kg/m, eight were trainee-performed and seven consultant-performed. Subcutaneous fat depth, muscle depth and total wound depth were associated with increased operative duration. BMI, abdominal girth, discectomy level and trainee-performed cases were not significantly associated with increased operative duration. Conclusions: Over half of the study patients were clinically obese. BMI and abdominal girth are not reliable predictors of operative difficulty as measured by operative duration. Lumbar discectomy in obese individuals is an appropriate training case. © 2018 Dallan P Dargan. Hosting by Science Repository. All rights reserved.


Introduction
An increasing proportion of the world's population are projected to suffer obesity and related diseases in coming decades. The majority of prolapsed lumbar discs can be managed non-operatively with success. For those requiring surgery, the next generation of spine surgeons must possess the surgical skills to treat high BMI individuals. Regarding the prolapsed intervertebral disc with associated sciatic nerve symptoms, surgery remains an established option for failed conservative treatment.
Body mass often requires adaptations of patient positioning and anaesthetic technique [1]. Higher complication rates after lumbar spine fusion in obese individuals have been observed [2]. In one singlesurgeon study of lumbar microdiscectomy in 75 patients, eight recurrences of herniated nucleus pulposus occurred. Among the 25 individuals with BMI >30kg/m 2 , the odds of recurrence were 12 times greater than their non-obese counterparts [3]. Some surgeons may conclude that lumbar discectomy is more technically demanding in high BMI individuals.
Duration of surgery has been used as a marker for technical difficulty, or unfamiliarity with the procedure. In the learning curve for lumbar micro endoscopic discectomy, the average operation duration reduces until approximately 20 cases [4]. Minimising operative duration remains crucial to managing theatre resources and lowering infection rates. Consequently, trainees may be directed towards non-obese patients as training cases. However, modern surgical instruments allow access to the lumbar disc at various wound depths while minimising soft tissue disruption. With regards to body mass index and body habitus, neither duration of surgery in lumbar discectomy, nor patient selection for training cases, have been examined.

Methods
A prospective observational study of elective primary lumbar discectomies was performed at two institutions by a single Consultant Orthopaedic Spine Surgeon (NWAE) and Specialty Registrar (CKJO). Pre-operative height and weight, abdominal girth in knee-chest position, and intra-operative wound measurements were taken. Operative duration was recorded using theatre management software.
All cases were performed during one six-month training rotation in orthopaedic spine surgery. Twenty-nine patients undergoing single-level lumbar discectomy were selected were selected, n=29, M:F ratio 18:11. Mean age was 47.2 years (range 18-83). The discectomy level was L4/5 in 18 cases, L5/S1 in 9 cases, with one case each at L2/3 and L3/4. BMI was calculated using pre-operative height and weight measurements. Patient positioning was standard knee-chest: prone with hips and knees flexed, and a vertical support against the buttocks and ischial tuberosities. The maximum abdominal girth was measured after induction of general anaesthesia, with the patient in the knee-chest position.
All cases underwent pre-operative image intensifier identification of spinal level, and pre-operative surgical incision skin marking. A standard minidiscectomy was performed in all cases. McCulloch or Caspar wound retractors were utilised. Wound depth, subcutaneous fat depth, and muscle depth were measured intra-operatively using a flexible sterile disposable ruler placed vertically in the wound. The deepest portion of the vertebral lamina was used as the base of the wound. Operative duration was recorded as the time from knife-to-skin until application of wound dressing. Patient positioning and level identification were not included in the operation time. Final wound length was measured. Whether wound extension was necessary during surgery to facilitate access was not recorded. The primary surgeon for each case was recorded prospectively in the patient clinical record.
A correlation was sought between the variables via Pearson's matrix. Correlation coefficient can be interpreted as follows: correlation coefficients (in absolute value) which are ≤0.35 are generally considered to represent low or weak correlations, 0.36 to 0.67 modest or moderate correlations, and 0.68 to 1.0 strong or high correlations with r coefficients ≥0.90 very high correlations [5]. Simple linear regression and stepwise linear regression with probability of rejection 0.05 was used to determine significant predictors of duration of the surgery. Analysis was carried out using Excel 2010 (Microsoft Corporation, Redmond, Washington) and Stata/IC 12.0 (StataCorp LP, College Station, Texas) and statistical significance was set at p≤0.05.

Discussion
Since Mixter and Barr's first description of the lumbar discectomy in 1934, alteration to the technique have been described [6]. The 'standard discectomy' incision from L3 to sacrum was used by Caspar to compare with microdiscectomy in 1991 [7]. Modern open discectomy involves intervertebral level identification via intraoperative imaging, and specialised tissue retraction devices. As a result, wound length and paraspinal muscle disruption are reduced considerably.
One series compares open 'macro discectomy' using a 5cm incision to microdiscectomy via a 2.5cm incision, with an operative time of 40 and 45 min respectively. Blood loss, and visual analogue scores for sciatica pre-and post-operatively were similar [8]. Operative duration for microdiscectomy, either using open technique or minimally invasive (paraspinal muscle dilatation without separation from spinous processes), have been compared, with duration of approximately 122 minutes [9]. A randomised controlled trial of open versus micro endoscopic discectomy, found longer incisions, longer hospital stays but shorter operative duration in the open group [10].
Regarding grade of surgeon, a retrospective audit of 971 primary elective lumbar microdiscectomies showed no association with outcomes [11]. Incidence of perioperative complications in thoracolumbar spine surgery was not shown to be significantly different among obese individuals, in a cohort of 87 patients, 40.8% of whom were obese [2]. Lumbar fusion in obese individuals with BMI >30kg/m 2 is associated with complication rates of over 40%, set-up times of one hour, and post-operative weight gain of 1.5kg versus 1.4kg weight loss in the morbidly obese (BMI >40kg/m 2 ) [12]. The perceived challenge of lumbar discectomy in obese patients may relate that of lumbar fusion, and to perioperative factors such as positioning of patients, and management of comorbidities rather than depth of wound. We observed that in patents with an elevated BMI or abdominal girth, a large proportion of the body fat was located within the anterior abdominal wall rather than posteriorly at the operative site. Therefore, we propose that trainees should not be denied opportunities in lumbar discectomy based on the patient's BMI or abdominal girth. Long-term follow-up for complications and validated outcome scoring is necessary to make detailed recommendations regarding patient selection for surgical training.

Conclusion
BMI, wound depth, and abdominal girth demonstrate weak correlations with operative duration. Total wound length has a significant association with operative duration in lumbar discectomy. A strong correlation exists between wound depth and subcutaneous fat depth. Cases performed by a consultant were shorter than those by a trainee in this study. The average duration of trainee-performed cases reduces over the course of fifteen cases. Elevated BMI and abdominal girth should not preclude surgical training in lumbar discectomy on the basis of anticipated prolongation of surgery.