Morbidity Following Hand-Assisted Laparoscopic Donor Nephrectomy

Introduction

In the United Kingdom the number of renal transplants performed continues to increase with 1413 deceased and 1043 living kidney transplants performed during the period 2016–2017. ¹ Living transplants offer a considerable advantage for the recipient with improved outcomes. ² The increased use of living transplantation has also reduced pressure on renal transplant waiting lists. However, living transplantation places the donor, an otherwise healthy individual, at risk of complications.

After the first successful laparoscopic donor nephrectomy (LDN) in 1995, ³ open donor nephrectomy (ODN) has become less favored. Most donor nephrectomies are now performed either as hand-assisted LDNs (HALDNs) or as purely LDNs. HALDN is a safe operation when compared with ODN with a similar rate of significant complications, as shown by Chandak et al. (1.4% cf. 1.8%). ⁴ However, patients undergoing HALDNs may have fewer minor complications than patients having an ODN, 16.7% compared with 37.5% in one study with patients able to return to regular activity sooner.^4,5

The reported overall rate of complications after HALDN is in the range of 13.5%–28%.^4,6,7 For example, Burkhalter et al. using Swiss registry data from 17 years found an overall morbidity rate after nephrectomy of 13.5%, although this was a heterogeneous group including open and laparoscopic cases. ⁶ More contemporary data, looking at HALDN only, collected between 2008 and 2012 by a Danish group, found an overall complication rate of 28% with 6 (6%) patients returning to the theater. ⁷ Mortality, after both ODN^8,9 and HALDN, is rare. ¹⁰

Given the popularity of HALDN and the importance of correctly appraising potential donors of the risks, it is essential to make sure these decisions are based on the best possible information. The authors, therefore, looked at operative and postoperative complications within our center to further inform the debate.

Methods

We retrospectively evaluated all HALDNs carried out within our center between August 2007 until June 2015. The donor details were obtained from a prospective database kept by the South West Transplant Centre. Patients with <6 weeks follow-up were excluded. Radiology data were collected from the patient archiving and communication system. Clinical data were collected from patients' clinical notes and pathology systems, as appropriate. Collected data included patient demographics, operative characteristics (including date of operation, intraoperative complications, and return to the theater), postoperative complications, and 30-day readmission data.

All nephrectomies were performed by 4 transplant surgeons, who were all experienced in donor nephrectomy surgery. Nephrectomy was performed through the transperitoneal approach in the lateral position. A hand port is inserted through a ∼8 cm horizontal supra- or infraumbilical incision for left and right nephrectomies, respectively. Aquagel^® was used as a lubricant for the hand port.

Statistical analysis was carried out using chi-square and Mann–Whitney U tests, where appropriate. A P value of <.05 was considered statistically significant. This study was undertaken following the Helsinki Declaration.

Results

During the study period, 161 patients had a HALDN with a median age of 51 years. There were 72 (44.7%) males, and 114 (70.8%) patients had left-sided surgery. The median length of stay was 3 days (range 1–8 days). Two (1.2%) cases were converted to open, both were left-sided nephrectomies and were discharged without complication. Complications were recorded in 21 (13%) individual patients. There were 25 separate 30-day readmissions during the study period representing 21 (13%) individuals (4 patients were readmitted twice). Table 1 compares the demographics, operative, and postoperative details of patients readmitted with those not readmitted.

The baseline data for both those readmitted and those not readmitted were broadly similar. Of the 21 patients readmitted, 9 (42.9%) were men, and the median readmission length of hospital stay was 3 days (range 2–8 days). Patients were readmitted at a median of 16 (range 0–30) days after nephrectomy. Only 1 patient was readmitted on the same day as discharge due to an abdominal wall hematoma requiring evacuation. Postoperative abdominal imaging was performed in 14 patients, with computerized tomography performed in 9 and ultrasonography performed in 6 patients. The most common findings on imaging were an intra-abdominal collection in 5 (36%) and nonspecific postoperative changes in 5 (36%) patients. Blood investigations, including inflammatory markers, were performed in 20 (95.2%) of 21 patients readmitted. The readmission diagnoses are shown in Figure 1.

OPEN IN VIEWER

FIG. 1.

Various causes of readmission after hand-assisted laparoscopic donor nephrectomy. In 3 patients who had appendicitis, inflammation was confined to the serosa of the appendix. AKI, acute kidney injury; DU, duodenal ulcer; OI, omental infarction; NSAP, nonspecific abdominal pain; NSTEMI, non-ST elevation myocardial infarction; UTI, urinary tract infection.

Five patients returned to the theater: 2 for appendicectomy, 1 for excision of an infarcted omentum and appendicectomy, 1 had repair of a perforated duodenal ulcer, and 1 patient had a laparoscopic washout of a turbid collection. All patients with appendicitis had serositis only, without mucosal inflammation on subsequent histology. All patients who underwent surgery either for appendicitis or for other indications had left-sided HALDNs. The side of nephrectomy and cause of readmission are given in Table 2.

Of the 9 patients without a definitive diagnosis, 8 had abdominal imaging on readmission. Imaging showed postsurgical changes in 5 patients and evidence of intra-abdominal collection in 3 patients. Eight of these patients were managed conservatively, and 1 patient had a laparoscopic washout. Seven of the 9 patients had raised inflammatory markers, with C-reactive protein (CRP) being most notable (Table 3). None of these patients had positive blood cultures and all these patients recovered without further sequelae.

Discussion

The main finding of this study was that a large proportion of readmitted patients had evidence of systemic inflammatory response (SIRS) but without an apparent cause. Although some studies have reported morbidity that is somewhat lower than this study group at ∼3.5%,^10,11 the morbidity rate in this study of 13% is comparable with other contemporary studies.^4,6,7 In particular, a recent retrospective analysis by Wiborg et al. reported an early complication rate of 28% ⁷ ; furthermore, review of >1600 nephrectomies including 569 HALDNs found a comparable complication rate (13.5%), ⁶ although readmission rate was not recorded.

Within this case series, there were several readmissions with nonspecific abdominal pain. In the majority of these patients, there were signs of SIRS. These patients represented 9 of 21 of the patients readmitted or ∼1 in 20 of the entire case series. All but 1 patient was readmitted >2 weeks after initial discharge with no specific cause found after imaging. Other groups have observed similar findings, for example, a case series of 150 HALDNs identified 30 readmitted patients who were considered to have “postnephrectomy inflammatory syndrome” with abdominal pain and/or fever, CRP >40, and a normal white cell count, ¹² the authors suggested this could be due to necrosis of perinephric fat left in situ after HALDN. ¹² Another study analyzed cytokine levels after 15 HALDNs but found no significant change from baseline in cytokine levels, suggesting only minimal systemic upset. ¹³ Although the underlying mechanism remains unclear, it is unlikely to be sepsis as none of these patients had positive blood cultures.

During HALDN, the only thing introduced into the peritoneal cavity is a lubricant gel (Aquagel^® [Parker Laboratories, Fairfield, NJ]), which comes into contact with the organs and peritoneal lining for an extended period of the operation. Lubricant gels are typically only used on intact skin, and even though the gel is sterile, it may cause a serositis in susceptible patients after a prolonged period of contact. For example, 3 patients developed appendix serositis during the postoperative period, suggesting an inflammatory process within the peritoneal cavity. We cannot explain how patients developed this; however, they probably had a sterile serositis. The finding supports this hypothesis that a similar gel caused an inflammatory reaction within the subarachnoid space in the animal model. ¹⁴ Our hypothesis is, therefore, that lubricant gel, such as that used in HALDN, can cause a serositis and systemic inflammation in susceptible patients.

Whatever may be the cause of this reaction, it represents an important and under-recognized cause of post-operative morbidity following HALDN. In the future, a change in practice may be necessary, such as using saline as a lubricant or moving toward a laparoscopic approach. However, further research is needed given our limited knowledge. After a review of our practice, the senior author (S.A.) has changed his practice, and now he only does LDN. We did not observe any readmissions in 20 LDNs that were performed for the past 24 months.

The two main limitations of this study are retrospective in nature and of small sample size. Despite these limitations, we observed a high rate of readmissions after HALDNs. In the majority of donors, we could not identify the cause for abdominal pain and intra-abdominal collection. Besides, we had donors who were readmitted with perforated duodenal ulcer and acute cholecystitis that are unrelated to the nephrectomy. We recommended that all potential donors who undergo HALDNs should be informed of the possible risk of readmissions due to nonspecific reasons and consented appropriately.