Perioperative outcomes of laparoscopic donor nephrectomy: A single-center experience

Abstract

Introduction:

Laparoscopic donor nephrectomy (LDN) remains the gold standard for living kidney donation. There is controversy over the optimal method for controlling the renal vessels. In this study, we assessed the outcomes of LDN and compared postoperative outcomes based on donor demographics and method of vascular control.

Materials and Methods:

We conducted a retrospective single-center review examining all adult patients who underwent LDN from 2017 to 2022. The outcomes collected included transfusion rate, estimated blood loss (EBL), reintervention rate, and overall 30-day complication rate. We also analyzed intra- and postoperative variables, including operative duration, conversion to open, method of vascular control, change in renal function, and length of stay.

Results:

168 patients were included, with the majority (81%) of patients undergoing left LDN. The median EBL was 100 mL (interquartile range 100–200 mL). Only two patients (1%) experienced hemorrhagic complications, though none was a result of clip or staple malfunction. The most common methods of arterial control were staples and clips (69%) and clips alone (28%), and the use of staples and clips was associated with longer operative duration. However, there were no significant differences in EBL, length of stay, or complication rate between groups. No patients required surgical reintervention for any reason, and there were no postoperative deaths.

Conclusion:

At our center, LDN was associated with low rates of hemorrhagic complications. There was no association between the method used to control the renal artery and perioperative outcomes or complication rates. Our data suggest that LDN is a safe procedure with low complication rates.

INTRODUCTION

Laparoscopic donor nephrectomy (LDN) is considered the gold standard approach for living kidney donation. Compared to the open approach, LDN is associated with reduced blood loss, shorter operative time, shorter warm ischemic time, improved cosmesis and postoperative pain, and faster recovery time.[1,2,3] One critical element of the procedure is the technique used to seal the renal vessels; these can be categorized as transfixating, in which material penetrates through the vessel wall, and nontransfixating, in which material is placed around the vessel without piercing the vascular tissue.[4,5]

The technique used can significantly impact the length of the donor vessels. The concerns over vessel length are particularly relevant to right-sided donor nephrectomy, as the right kidney has a shorter renal vein and thus is less preferable for donation than the left kidney.[6] Staples generally result in decreased vessel length, as multiple rows of staples must be trimmed from the vessel prior to anastomosis.[7,8] However, when compared to clips, stapling is widely considered to be the safer and more reliable technique.[4,9] Both methods also carry a risk of failure, either due to stapler misfire or clip slippage.[8,10] As a result, the technique utilized must balance the need to achieve appropriate vessel length for the transplant anastomosis with the risk of device failure, which can result in catastrophic bleeding.[9]

Studies comparing the safety of clips to staples have reported no significant differences in complication rates between the two methods.[8,10,11,12] Despite the perception that clips are inferior to staples when accounting for device safety and complication rates, they remain widely used in LDN and their relative safety compared to staples remains uncertain. In this study, we aimed to perform a retrospective chart review comparing different methods of vascular control and evaluating perioperative outcomes of LDN at our centre, focusing on hemorrhagic complications.

MATERIALS AND METHODS

Study design

Following approval from our institutional research ethics board, we conducted a retrospective chart review at a single high volume renal transplant center, including all patients aged 18 years or older who underwent LDN between January 1, 2017, and April 30, 2022. Data collected included patient characteristics (age, sex, body mass index [BMI], comorbidities, surgical history, and American Society of Anesthesiologists [ASA] classification), preoperative and postoperative laboratory values (creatinine, estimated glomerular filtration rate [eGFR], and hemoglobin), intraoperative variables (surgical approach, donor kidney laterality, number of arteries and veins, method of vascular control, estimated blood loss (EBL), and conversion to open), and postoperative outcomes (operative duration, 30-day complication rate, transfusion rate, repeat procedures, and duration of hospital stay). The difference in hemoglobin and renal function was assessed by comparing the most recent preoperative hemoglobin, creatinine, and eGFR to the same laboratory values collected on the date of discharge. Complications were categorized based on the Clavien–Dindo classification system. Postoperative outcomes were compared between between method of vascular control.

Surgical technique

LDN was performed using a standard transabdominal approach by three experienced transplant surgeons. Pneumoperitoneum was achieved through insufflation through a Veress needle. Endovascular stapling was performed using either a Covidien Endo GIA cutting stapler or an Endo TA non-cutting stapler; with the TA stapler, the vessel was transected after stapling using laparoscopic scissors. Clipping was performed using titanium clips, applied with laparoscopic clip appliers. Hem-o-Lok clips are not used for any LDN cases at our center. The method used to secure the renal vessels was determined by the surgeon preference.

For some right-sided donor nephrectomies, a laparoscopic-assisted approach was utilized in order to maximize vessel length. In this approach, a subcostal incision was used to open the abdomen after the renal vessels had been exposed and the kidney had been fully mobilized. A Satinsky clamp was then applied to the inferior vena cava (IVC), and the renal vein was divided along with a cuff of IVC, thus ensuring maximum vessel length. The caval defect was oversewn with sutures, and the kidney was removed. The decision to proceed with this approach rather than a standard laparoscopic approach was based on surgeon preference.

Statistical analysis

Data extracted from patient records were collected with Microsoft Excel 365 (Microsoft, Redmond, Washington) and analyzed using IBM SPSS v28 (Armonk, United States: IBM Corp). P < 0.05 was considered statistically significant. The normality of data was determined using the Shapiro–Wilk test. Continuous data were presented as means with standard deviations for data that followed a normal distribution or as medians with interquartile ranges (IQRs) for data that did not. Dichotomous or categorical data were presented as proportions. Differences between groups were evaluated using the Chi-square or Fisher’s exact tests for categorical data and Student’s t-tests or Mann–Whitney U tests for continuous data. Missing data were excluded from analysis in a pairwise fashion.

RESULTS

Patient demographics and perioperative outcomes

A total of 168 patients underwent LDN at our center between January 1, 2017, and April 30, 2022. Patient demographic information is summarized in Table 1. 102 (61%) of the included patients were female, and the median age of patients undergoing LDN was 51 years (IQR, 39–59 years). 153 (91%) donors were healthy with no significant comorbidities, and the mean BMI of included patients was 27 kg/m² ± 4.0kg/m². 153 (91%) patients were classified as ASA I or II.

Table 1.

Demographic information

Variable	n (%)
Patients	168
Male	66 (39)
Female	102 (61)
Age (years), median (IQR)	51 (39–59)
BMI (kg/m2), mean±SD	27 (4.0)
ASA
I	67 (40)
II	86 (51)
III	15 (9)
Comorbidities
None	153 (91)
Hypertension	12 (7)
Cardiac	3 (2)

IQR: Interquartile range, BMI: Body mass index, SD: Standard deviation, ASA: American Society of Anesthesiologists

Operative characteristics

During our study period, 136 (81%) patients underwent left LDN, whereas 32 (19%) donated their right kidney. Of these right-sided donors, 20 (63%) underwent a laparoscopic-assisted nephrectomy with planned opening through a subcostal incision. The majority of patients had one renal artery (84%) and one renal vein (96%). The median operative duration was 201 min (IQR, 177–240 min), and the median EBL was 100 mL (IQR, 100–200 mL). Intraoperative and postoperative data are summarized in Table 2.

Table 2.

Intraoperative and postoperative variables and outcomes

Variable	n (%)
Laterality of donor kidney
Left	136 (81)
Right	32 (19)
Number of renal arteries
One	141 (84)
Two	27 (16)
Number of renal veins
One	161 (96)
Two	7 (4)
Surgical approach
Pure laparoscopic	148 (88)
Hand-assisted	20 (12)
Conversion to open	2 (1)
Aborted	2 (1)
Operative duration (min), median (IQR)	201 (177–240)
EBL (mL), median (IQR)	100 (100–200)
Hb decrease (g/L), median (IQR)	19 (15–24)
Creatinine increase (μmol/L), median (IQR)	32 (23–41)
eGFR decrease (mL/min/1.73 m2), median (IQR)	31 (11)
LoS (days), median (IQR)	3 (2–3)

IQR: Interquartile range, eGFR: Estimated glomerular filtration level, Hb: Hemoglobin, EBL: Estimated blood loss, LoS: Length of stay

Postoperative outcomes and complications

At the time of discharge, the median decrease in hemoglobin was 19 g/L (IQR, 15–24 g/L). In terms of renal function, the median increase in creatinine was 32 umol/L (IQR, 23–41 umol/L), and the median decrease in eGFR was 32 mL/min/1.73 m² (IQR, 24–40 mL/min/1.73 m²). The median length of stay was 3 days (IQR, 2–3 days).

Age ≥65 years and BMI ≥25 kg/m² did not impact perioperative outcomes [Table 3]. However, female sex was associated with a lower median postoperative eGFR decrease compared to male patients (P = 0.01). The presence of comorbidities was associated with a significantly longer median operative duration (P = 0.043). Right-sided kidney donation was associated with a significantly higher hemoglobin drop (P = 0.046) compared to left-sided donors. When comparing left-versus right-sided donors, there was a significant difference in eGFR decrease (P = 0.006) and length of stay (P = 0.007), but the median length of stay and eGFR decrease were equivalent between groups.

Table 3.

Relationship between patient demographic and clinical factors and perioperative outcomes

	P
	EBL	Hb decrease	eGFR decrease	LoS	OR duration
Age ≥65 years	NS	NS	NS	NS	NS
BMI ≥25 kg/m2	NS	NS	NS	NS	NS
Presence of comorbidities	NS	NS	NS	NS	0.043
Female sex	NS	NS	0.046	NS	NS
Right-sided donor	NS	0.046	0.006*	0.007*	NS

*No difference in median values between groups. EBL: Estimated blood loss, Hb: Hemoglobin, eGFR: Estimated glomerular filtration rate, LoS: Length of stay, OR: Operative, NS: Not significant, BMI: Body mass index

Vascular control

The most common method of controlling the renal artery was with both staples and clips, which was used in 116 (69%) cases. Clips alone were used in 47 (28%) cases, and staples alone were used in three (2%) cases. In regard to the renal vein, 139 (83%) of cases involved using staples alone, six (4%) cases involved clips alone and three (2%) cases involved using both staples and clips. 19 (11%) cases involved directly suturing the IVC, with all such cases being right-sided laparoscopic-assisted donor nephrectomies.

To compare different methods of vascular control, we compared outcomes between patients whose renal artery was controlled with both staples and clips to patients whose artery was controlled with clips alone; these two methods were chosen as they were the two most common methods of controlling the renal artery. There was no significant difference between groups in the EBL, length of stay, hemoglobin decrease, creatinine increase, eGFR increase, or complication rate. However, we found that the median operative duration was significantly longer when both staples and clips were used to control the renal artery compared to with the use of clips alone (P = 0.008); the median operative duration was 192 min (IQR, 160–215 min) with clips alone compared to 204 min (IQR, 189–249 min) with both staples and clips.

Complications

Of the 168 included donors, three experienced complications that were categorized as Clavien–Dindo Grade III or higher. In the first case, the donor experienced a massive intraoperative bleed when the Satinsky clamp slipped off of the IVC during laparoscopic-assisted right donor nephrectomy. The patient became pulseless from profound blood loss, but the return of systemic circulation was quickly achieved with resuscitation. The vascular surgery service was consulted, and the caval defect was repaired. Intraoperative blood loss was 3L, and the patient received 5 units of packed red blood cells (pRBCs) as well as 4 units of fresh frozen plasma.

The donor was admitted postoperatively to the intensive care unit (ICU), where she was stabilized and ultimately transferred to the ward after one day. On postoperative day six, she was diagnosed with a small asymptomatic IVC thrombus but was cleared for discharge without anticoagulation by the thrombosis service. She returned to the emergency department six days later with the thrombus now extending down to both common iliac veins, resulting in occlusion of the left common iliac vein and partial occlusion of the right common iliac vein. She was admitted to the hospital and treated with heparin, ultimately being discharged on tinzaparin after a subsequently uneventful hospital stay. Follow-up Doppler ultrasound showed improvement in her deep vein thrombosis (DVT) and return of normal flow to the common iliac vessels. Ultimately, the donor did not develop any long-lasting deficits related to the donor nephrectomy or her DVT.

In the second case, the patient developed a right pneumothorax in the recovery unit, likely related to the patient’s preoperative intercostal nerve block. The patient was assessed by the thoracic surgery service, and a pigtail chest drain was placed at the bedside. By the next day, the pneumothorax had resolved, and the chest drain was removed. The remainder of the patient’s postoperative course was unremarkable.

In the third case, the patient developed persistent postoperative sinus tachycardia that did not respond to intravenous fluids or beta blockers. Although she was normotensive and asymptomatic, she was admitted to the ICU postoperatively for cardiac monitoring and given 1 unit of pRBCs empirically. Computed tomography confirmed that there was no pulmonary embolism, and her heart rate improved over the course of several hours with additional intravenous fluids and oral beta blockers. After she was transferred back to the postoperative ward, her hospital course was unremarkable. Her tachycardia was attributed to hypovolemia and anxiety, and she was ultimately discharged with no issues.

Aside from the patients described above, no other patients required blood transfusion, ICU admission, or additional procedures. There were no postoperative deaths within 30 days of donor nephrectomy.

Two patients required unplanned conversion to open nephrectomy. In the first case, a Gibson incision was made due to difficulties dissecting the perinephric fat laparoscopically; after conversion to open nephrectomy, the remainder of the procedure was uncomplicated. In the second case, the decision to convert was made due to persistent bleeding from the gonadal vein that was difficult to control; after opening, the vessel was suture ligated, and the nephrectomy was completed successfully without any further significant blood loss.

Two cases were aborted intraoperatively. In the first case, the donor had already been induced under general anesthesia when the recipient was found to have elevated troponin levels during preoperative evaluation. Given this finding, both surgeries were canceled, and the donor was awoken from anesthesia before an incision was made. In the second case, a lower pole artery was damaged during the dissection of the renal vessels, and the artery ultimately needed to be sacrificed. Based on this, the decision was made to abort the nephrectomy due to concerns regarding impaired renal function secondary to the partial devascularization of the donor kidney.

DISCUSSION

The safety of LDN has been well-established in the literature, with the procedure demonstrating low rates of perioperative complications and good long-term donor outcomes.[13,14,15,16,17,18,19] In our single center retrospective chart review, only three patients experienced complications that were classified as Clavien–Dindo Grade III or higher, and there were no deaths or reoperations. Two patients received blood transfusions, and only one was transfused for confirmed hemorrhage. In addition, two cases were aborted; one case was aborted due to anesthetic concerns regarding an elevated preoperative troponin, and the other due to concerns of renal devascularization after the sacrificing of an accessory renal artery. Two patients required conversion to open, but neither case was due to bleeding. Our overall 1.8% overall major complication rate and 1.2% transfusion rate are consistent with other studies examining the outcomes of LDN in Canada.[11,20]

This low complication rate is bolstered by the stringent selection criteria for LDN, which selects for donors that are good surgical candidates with no major comorbidities.[21] Indeed, the Canadian review of the international Kidney Disease: Improving Global Outcomes (KDIGO) guidelines for living donors recommends selecting for donors with good preoperative renal function and minimal risk factors for developing renal, cardiovascular, oncologic, and infectious disease.[22] Accordingly, donors in our cohort were typically young, with a median age of 51 years. Only 9% of patients had any significant medical comorbidities, and the median BMI was 27 kg/m². From an anesthesia perspective, 91% of patients in our study were classified as ASA I or II, indicating a low risk of anesthetic complications.

Demographic factors, including age ≥65 years and BMI ≥25 kg/m² did not impact perioperative outcomes. While significant differences in outcomes were found when comparing patient sex, they were unlikely to be of any clinical significance. For example, while female patients had a lower postoperative decrease in eGFR when compared to males, the median difference was only 4 mL/min/1.73 m². Similarly, the presence of comorbidities was associated with an increase in median operative duration of 21.5 min. Finally, right kidney donation was associated with a relatively small 2 g/L improvement in hemoglobin drop when compared to left-sided donors. Overall, these differences, while statistically significant, are unlikely to have any substantial impact on perioperative outcomes and should not influence patient selection or surgical planning.

Much of the analysis of LDN safety has focused on the method of vascular control during LDN.[10] Multiple surveys of transplant surgeons in Canada, the United States, and Europe have found that the majority of surgeons prefer the use of staples during LDN, likely due to their perceived superior safety profile compared to clips.[4,9,23,24] While stapler malfunction is rare, a misfire can result in an incomplete staple line, leading to bleeding from the renal vessel following transection.[25] In a 2021 review of the Food and Drug Administration (FDA) Manufacturer and User Facility Device Experience database, Gopal et al. identified 383 cases involving complications associated with endovascular staplers during laparoscopic nephrectomy from 2009 to 2019, with 22 deaths attributed to stapler misfire.[26]

The safety of clips has also been called into question despite their ability to provide superior vessel length in the donor kidney.[12,27] In 2006, following multiple cases of clip slippage resulting in donor death, the FDA initiated a Class II recall of Hem-o-lok clips and subsequently issued safety information in 2011 that Hem-o-lok clips are not recommended for use in donor nephrectomies.[28,29] Despite this perception, several studies have reported relatively low rates of complications, including hemorrhage and death, when using polymer clips alone.[20,30,31,32,33] At our center, we do not use Hem-o-Lok clips for LDN, instead using only titanium clips.

Some surgeons have suggested that, when compared to Hem-o-lok clips, titanium clips may be even more prone to slippage, as they lack a locking mechanism.[10,34,35] However, multiple studies have found low rates of device failure and subsequent complications with metal clips.[35,36,37] In addition, given their relatively simple design, titanium clips may be more cost-effective than either Hem-o-lok clips or endoscopic staples.[35,37] Meta-analyses assessing the safety of various vascular closure devices during LDN have found no difference in hemorrhagic complications, device failure, and perioperative death when comparing vascular staples, polymer clips, and titanium clips.[8,38,39] Given the lower cost of both titanium and polymer clips when compared to staples, as well as the increased vessel length that clips provide, the use of clips during LDN may provide economic benefits without compromising donor outcomes.[8,37,39]

In our cohort, the two most common techniques for securing the renal artery were titanium clips alone or a combination of both staples and clips. Comparing these methods, the only significant difference we found was that the median operative duration was approximately 12 min shorter when using clips alone. Such a small difference in operative duration is unlikely to have a significant impact on the clinical outcomes or logistics of LDN. Due to our low overall complication rate, we could not accurately compare rates of perioperative complications based on the method of vascular control. Based on our experience, we have found that both endovascular staples and titanium clips, used individually or in combination, are safe and effective techniques for controlling the renal vessels during LDN. Therefore, we recommend that surgeons can safely utilize clips or staples for vascular control during LDN based on their personal or institutional preferences.

Our study is not without its limitations. As our patient cohort was retrospectively obtained from a single high-volume center, the generalizability of our results to the broader population of kidney donors or lower-volume centers is unclear. In addition, since our data was extracted from information documented in operative notes and patient charts, it is possible that some data related to operative factors and donor outcomes were missed. As previously mentioned, only patients who are relatively healthy are considered for kidney donation, which introduces an inherent source of bias. In addition, since patients were not prospectively divided into groups based on the technique used to secure the renal vessels, our ability to compare outcomes between these techniques may be limited. Furthermore, since the vast majority of cases involved controlling the renal vein with staples alone, we did not have the statistical power to make comparisons between different methods of securing the renal vein. Finally, as we focused on immediate perioperative outcomes within 30 days of surgery, the assessment of long-term outcomes is beyond the scope of this study.

CONCLUSION

LDN is a safe procedure with low complication rates and perioperative morbidity. In appropriately selected donors, demographic factors such as age, BMI, and comorbidities do not have a significant clinical impact on perioperative outcomes after LDN. At our center, there were low rates of hemorrhagic complications following LDN. In addition, no perioperative complications were associated with a stapler or clip malfunction. When comparing different methods of vascular control, the use of staples and clips resulted in longer operative duration compared to using clips alone. However, overall perioperative outcomes were favorable regardless of the method used to secure the renal vessels. Further prospective comparative studies, including randomized controlled trials, would shed more light on the relative safety of different methods of vascular control.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.