Lower limb compartment syndrome as a complication of robot-assisted radical prostatectomy: the UK experience

Authors


Correspondence: Conrad V. Bishop, Unit 3, 28 Carlingford Road, London NW3 1RX, UK.

e-mail: conradbishop@hotmail.com

Abstract

Objective

  • To evaluate the incidence of lower limb compartment syndrome (LLCS) in robot-assisted radical prostatectomy (RARP) and the prevalence of risk factors in patients with LLCS.

Methods

  • Data were collected from 17 UK robotic surgery institutions for a multicentre analysis.
  • Data were used to calculate the incidence of LLCS and the prevalence of risk factors.

Results

  • A total of 3110 RARPs were performed by 17 institutions between 2004 and 2011.
  • There were nine cases of LLCS, giving an incidence of 0.29%. Seven of these required fasciotomy.
  • The prevalence of risk factors was as follows: console time >4 h in 8/9 cases, early learning curve (<20 cases) in 3/9 cases; obesity (BMI >30 kg/m2) in 5/9 cases; and peripheral vascular disease in 2/9 cases. One patient with LLCS was positioned incorrectly.

Conclusions

  • The serious complication LLCS occurs in RARP but has a low incidence.
  • Long operating times, surgical inexperience, poor patient positioning, obesity and vascular disease appear to be risk factors.

Introduction

Lower limb compartment syndrome (LLCS) is a rare but serious complication of surgery in the lithotomy position [1-5], with an estimated incidence of one in 3500 cases [4]. The resulting injury often leads to disfigurement, amputation and permanent disability and it is fatal in an estimated 6% of cases [2]. LLCS in robot-assisted radical prostatectomy (RARP) has been described in isolated case reports [6-8], but it does not feature in large multicentre reviews of complications of RARP [9, 10] and therefore the exact incidence of LLCS in RARP is unknown. Many underlying risk factors for LLCS, aside from surgery in the lithotomy position, have been postulated. Consistent risk factors in the literature include the Trendelenberg position, prolonged operating time, obesity, peripheral vascular disease and external calf compression [1-5]. While not postulated as a risk factor for LLCS in the published literature, we believe that surgical inexperience may contribute to poor patient positioning and increased operating times, both of which are recognized to increase the risk of LLCS [1-5]. The primary aim of the present study was to be the first to estimate the incidence of LLCS in RARP as well as identify any associated risk factors in this group. A secondary aim was to draw attention to this serious complication and suggest recommendations for its prevention.

Material and Methods

Urology units within the UK with an established RARP programme (defined as >30 RARP cases) were invited to contribute data for a multicentre analysis. These retrospective data were acquired on the condition of anonymity of the participating units to encourage an open and frank assessment of a major complication. Data included the RARP caseload, incidence of LLCS, subsequent management of LLCS and assessment of both surgical and patient risk factors. Surgical risk factors included long operating times (console time >4 h), early learning curve of the surgeon (<20 cases) and the use of pneumatic compression stockings. Patient risk factors included obesity (BMI >30 kg/m2) and a history of symptomatic peripheral vascular disease. The incidence of LLCS was then calculated for the cohort, as was the prevalence of risk factors in patients with LLCS. These figures formed the basis for conclusions to be drawn and recommendations made for future practice.

Results

Seventeen of 22 urology units in the UK with a robotic surgery programme elected to participate in this study. A total of 3110 RARPs were performed by these institutions from November 2004 to February 2011. The RARP caseload among institutions ranged from 40 to 850 cases. The breakdown of this caseload is shown in Table 1.

Table 1. Number of RARPs carried out in participating UK institutions vs incidence of LLCS
RARP caseload, nUK institutions, nLLCS episodes, n
  1. *Denotes two cases of LLCS from a single centre.
<10053*
100–20084*
200–50021
>50021

Nine cases of LLCS were reported from seven centres, giving an incidence of LLCS in this RARP cohort of 0.29%. Two centres reported two cases of LLSC.

All patients with LLCS after RARP had at least one postulated risk factor. The prevalence of risk factors in patients with LLCS is shown in Table 2. Of the two patients with known peripheral vascular disease, one had previously undergone vascular bypass grafting, while the other had mild symptoms which were treated conservatively. One surgeon, early in his learning curve, volunteered the information that he had failed to properly position the patient, probably contributing to a case of LLCS. All patients in the series wore pneumatic compressive stockings throughout the procedure.

Table 2. Prevalence of risk factors in cases of LLCS
Patient no.ObesityPeripheral vascular diseaseConsole time >4 hEarly learning curvea
  1. a<20 cases.
1  YesYes
2 YesYesYes
3Yes Yes 
4Yes Yes 
5  YesYes
6 Yes  
7Yes Yes 
8Yes Yes 
9Yes Yes 
Total5283

All cases of LLCS were assessed by either a vascular or orthopaedic surgery team. The timing of this assessment was not volunteered in any case. Invasive measurement of compartment pressures was not performed for diagnosis or follow-up in any case. Seven patients underwent fasciotomy. Five patients later underwent primary closure of the fasciotomy, while two patients required skin grafting to cover the defect. Two patients were treated conservatively with i.v. fluids and analgesia. While these two patients did not have compartment pressures measured, it was the opinion of the specialist team that a clinical diagnosis was sufficient as all other causes of pain had been excluded. There were no amputations or deaths from LLCS in this series. There were no reports of rhabdomyolysis or subsequent complications such as renal failure or hyperkalaemia.

Discussion

Lower limb compartment syndrome is a devastating condition produced by a cascade of biochemical, physiological and pathological responses within the fascial compartment of the limb. In the case of surgery in the lithotomy position, these responses are triggered by a decrease in blood flow as the patient's legs are above the level of the heart for a prolonged period. This decrease in blood flow can be exacerbated by direct pressure from the calf support [11]. The hypoxia within the compartment leads to an inflammatory response and an associated release of cytokines, leading to capillary leakage and interstitial oedema. This in turn further increases the pressure within the compartment, escalating into a self-perpetuating cascade of worsening hypoxia and swelling. As the compartment pressure rises above critical levels, nerves, blood vessels and muscle become affected. Nerve compression initially causes severe pain and later neuropraxia as the condition worsens. Pressure on the vessels further reduces blood flow causing limb ischaemia. A combination of direct pressure and ischaemia causes myonecrosis, which can lead to rhabdomyolosis, cell destruction, renal failure, hyperkalaemia and death [2, 5].

Aside from the lithotomy position, placing the patient head down in the Trendelenberg position further exacerbates this decrease in lower limb blood flow by increasing the vertical distance between the heart and lower limbs. Studies with invasive monitoring show that a Trendelenberg tilt by as little as 15° is enough to cause a significant change in compartment pressure [12].

Correct positioning of the patient is paramount in preventing LLCS [1-5, 11]. In our series, one surgeon admitted failure to recognize a poorly positioned patient who subsequently developed LLCS. We recommend that all surgeons check the position of the legs before commencing the procedure (see summary of recommendations in Table 3). Ideally, the weight of the leg should be supported on the patient's heel through the stirrup, with the upper part of the calf clear of the calf support. A simple test is to slide a hand between the support and the calf to check for a gap.

Table 3. Summary of recommendations to prevent and manage LLCS
Recommendations
  1. Active mentorship during the learning curve to keep console time <4 h.
  2. Careful case selection during the learning curve.
  3. Check for correct positioning of the legs before commencing procedure.
  4. Replace the legs to a level position immediately after the robot has undocked.
  5. Have a high index of suspicion in patients with leg pain in recovery.
  6. Early specialist referral for suspected cases.

Console times >4 h appear to be the dominant risk factor for LLCS in this series, occuring in 8/9 cases. It would be interesting to know the exact operating times in these cases to see how much in excess of 4 h they were, however, owing to the anonymous nature of the data collected, we had no way of identifying the individual cases in order to look up the exact operating times. In light of this relationship with time, we recommend removing the legs from the calf support to the supine position as soon as the robot is undocked to decrease the time spent in the lithotomy position, rather than leaving the legs up while extracting the specimen, placing a drain and closing the wounds.

Surgeons in their learning curve were prevalent in incidences of LLCS in this series. The relationship between longer operating times and the learning curve of RARP has been previously described in the UK [13, 14]. We believe good mentorship is vital to safely assist a surgeon through the learning curve. When providing mentorship to a colleague, there is great pressure on the mentor to maximize the console time of the colleague to allow that person to complete as much of the case as possible. We feel it is the mentor's responsibility to keep the case to a reasonable timeframe in the interest of patient safety and that the mentor needs to be proactive in taking over the more challenging steps to keep the console time to <4 h. Based on our data, it is reasonable to conclude a mentor should be present for all cases until a surgeon can demonstrate the skills required to complete a case within 4 h, regardless of how many cases are required to achieve this. During this period, case selection is important and patients with previous abdominal surgery, very large or small prostates or those requiring extended pelvic node dissection are best avoided to keep the operating time to a minimum.

Surgical inexperience may also be a factor in poor positioning of the patient; it is important that the mentor does not overlook providing guidance in positional checks at the beginning of the procedure.

Obesity and peripheral vascular disease were both prevalent risk factors for LLCS. A major drawback of this series was that the prevalence of these risk factors was not known for the entire study population, thus risk ratios and statistical significance could not be calculated. According to the Hertfordshire and South Bedfordshire Urological Cancer Centre's prospective RARP database, currently consisting of 227 RARPs, the prevalence of obesity and peripheral vascular disease is 3.5 and 2.2%, respectively, compared with 55.6 and 22.2%, respectively, in patients with LLCS across the whole cohort. Assuming this prevalence were uniform across the RARP population of all 17 institutions, the increased prevalence of both risk factors would reach significance, with P < 0.001 for obesity and P < 0.05 for peripheral vascular disease. While we concede the prevalence of these risk factors is likely to vary throughout the UK RARP population through differences in demographics and case selection, it would be difficult to mount a credible argument that these risk factors are not associated with the incidence of LLCS.

While statistical significance is an important tool to correlate obesity and peripheral vascular disease with LLCS, the calculation of risk ratios is of less value owing to the heterogenous nature of both factors. As such, a single risk ratio for either factor would be of little use in practical terms. For example, a patient with a BMI of 35 kg/m2 would have a different risk from that of a patient with a BMI of 30 kg/m2, and two patients with the same BMI but with different body shapes and fat distributions would have different risks for LLCS. Similarly, a patient with diabetes and small vessel disease in the lower limb would have a different risk for LLCS from that of a patient with an endovascular stent for a large vessel atheroma. Regardless of how these risks may vary in individuals, the point remains that the risk is increased and caution must be exercised. We recommend surgeons avoid patients with these risk factors during the learning curve.

As calf compression is a known risk factor for LLCS [1-5], this brings into question the use of pneumatic compression stockings during RARP. These devices, largely the Flowtron® Intermittent Pneumatic Compression stockings (ArjoHuntleigh AB, Eslov, Sweden), were used across the study population so we were unable to determine any significance of this as an independent risk factor for LLCS. Whether or not pneumatic compression stockings contribute to LLCS, the fact remains that thromboembolic events are also a serious and sometimes fatal complication of RARP. The rates of thromboembolism after RARP vary between 0.2 and 1.5% in the world literature [9], mostly higher than our incidence of LLCS; therefore, we have no basis to recommend the discontinuation of pneumatic compressive stocking use during RARP.

In recognizing compartment syndrome, medical school teaches us to look for the six classic ‘Ps’, pain, pressure, pallor, paralysis, paraesthesia and pulselessness. Pain is the most reliable sign [15], and in this context, is first demonstrated when the patient wakes in recovery. The pain is typically not adequately controlled with narcotic analgesia. The other ‘Ps’ manifest later in the progression of the cascade, and while worth checking for, their absence does not exclude LLCS [15]. As outcomes from decompressive fasciotomy become poorer with increasing time [15], a high index of suspicion is required in patients with leg pain in recovery and urgent specialist attention should be sought where there is concern. Invasive measurement of compartment pressures may aid the diagnosis where a clinical diagnosis in inconclusive, but interpretation of the pressures may be difficult depending on the device used [16]. We feel this is out of the scope of expertise of the urologist and best left to a specialist in the field.

In conclusion, the present study draws attention to a rare but debilitating complication of RARP. We believe it is largely preventable with sensible mentorship, an understanding of correct patient positioning and careful assessment of patient risk factors. We hope that an awareness of LLCS will lead to early recognition and prompt intervention, should the complication arise.

Conflict of Interest

None declared.

Abbreviations
LLCS

lower limb compartment syndrome

RARP

robot-assisted radical prostatectomy

Ancillary