Patient selection. From May 1995 to December 1997 living donor nephrectomy was performed on sixty-two consecutive patients at the Milton S. Hershey Medical Center, Hershey, Pennsylvania. Thirty patients underwent nephrectomy using the conventional open retroperitoneal approach (Group B) and next thirty-two nephrectomies were performed using the minimally invasive posterior supracostal or modified Turner-Warwick approach (Group A),

All potential donors were identified on the basis of ABO compatibility, HLA (human leukocyte antigen) tissue typing and final leukocyte crossmatching. A thorough medical evaluation was performed prior to surgery including blood chemistries, creatinine clearance, electrocardiogram and chest radiograph. An oral glucose tolerance test was performed in patients with a family history of diabetes. Prior to surgery all donors underwent arteriography to define vascular anatomy. When multiple vessels were present on one side the contralateral kidney was chosen. The left kidney was taken preferentially when all factors were equal.

Data collection. Preoperative donor data, including gender, age, body weight and number of arteries were obtained from a computerized database. The medical record of each patient was reviewed to assess for perioperative and postoperative variables: including operative time, estimated blood loss, time to resumption of oral intake, transfusion requirement, and length of hospital stay. In hospital postoperative analgesic requirements, parenteral and oral, were calculated first in mg of morphine sulfate (MSO4) (10 mg morphine sulfate = 1 morphine sulfate equivalent) using the conversion chart (15). The length of the incision was measured at the first clinic visit approximately one month following surgery. Information regarding post-operative return to full activity was collected prospectively in both conventional and minimally invasive approach via the telephone by a registered nurse who was blinded to the type of donor nephrectomy.

Statistics. Statistical analyses were performed using a group mean with the standard error. Unpaired t test was used to analyze the differences between groups using the Prism statistical software package. Statistical significance was reached with p values < 0.05.

Operative techniques. The surgical approach to the conventional open retroperitoneal nephrectomy has been well described previously (13,16,17). In 1965, R.T. Turner Warwick first proposed the posterior supracostal approach to the renal area (14). This approach was advocated by the author as a means for better lateral exposure of the kidney and easier wound closure (Figure 1). The approach took advantage of the not widely appreciated fact that the eleventh and twelfth ribs have a single articulation with the vertebral column. If the superior ligamentous attachment is severed, either rib can be hinged inferiorly to give maximal exposure of the renal bed (Figure 2).

Following induction of general endotracheal anesthesia, a Foley catheter is placed to monitor urine output throughout the case. The patient is placed in the standard lateral decubitus position. The arteriogram has been previously examined to determine the level of the kidney and the eleventh or twelfth rib is chosen for the initial incision.

The incision begins at the tip of the eleventh or twelfth rib and extends posteriorly over the rib for a distance of 10-14 cm (Figure 3). The subcutaneous tissue is divided by electrocautery and the attachments of the latissimus dorsi and serratus anterior are divided for a short distance along the rib. The superior attachments of the intercostal muscles to the rib are divided by electrocautery. The ligaments to the vertebral column posteriorly are divided bluntly and the rib is hinged inferiorly (Figure 1).

To maximize the size of the operative field the diaphragmatic attachments to the rib are taken down using electrocautery and exposure to the field is maintained using a hand-held retractor (in case of conventional open nephrectomy, Finochietto retractor, Kapp Surgical Instrument, Inc. Cleveland, Ohio). Gerota's fascia is sharply incised and its attachments to the kidney are taken down. The superior pole of the kidney is dissected free from the surrounding tissue first, followed by the inferior pole.

The ureter is clearly identified and dissected to the level of iliacs. Two large clips are placed on the ureter distally and the proximal portion is brought up into the operative field. The arterial and venous blood supplies of the donor kidney are dissected free from the surrounding tissue. Branches to renal vein and from the artery are clipped twice proximally and distally, rather than tied, because the short incision precludes safe ties by hand.

At the beginning of the hilar dissection, 0.5 mg/kg (body weight) of mannitol and 20 mg of furosemide are administered intravenously tot the donor. Three minutes prior to removing the kidney 50-60 units per kilogram of heparin is given intravenously. Warm ischemia time with either approach is measured in seconds as the kidney is flushed on the prepared back table with 300 - 400 ccs of Eurocollins solution immediately after removal. After completion of the nephrectomy the previously retracted rib is allowed to assume its natural position. The muscles are closed in two layers. The subcutaneous tissue is approximated and the skin is closed with a subcuticular suture.

Results

From May 1995 to December 1997 a total of sixty-two consecutive donor nephrectomies were performed: the first thirty donors underwent conventional open donor nephrectomy (Group B), and the next thirty-two were performed using the minimally invasive supracostal approach (Group A). Patient weight or size was not a contraindication to surgery in either group. There were no vascular or ureteral injuries to any of the donor kidneys. All kidneys were transplanted into their potential recipients and functioned immediately. The demographic characteristics of the donors are shown in Table 1. There were no significant differences in the gender, race, age, or body weight between the two groups.

The variables associated with the intraoperative and postoperative course of the two groups are shown in Table 2. Operating time was measured from the initial surgical incision to final closure. There was no statistical difference in operating times between Group A, 176.8 ± 11.2 minutes, versus Group B, 178.6 ± 7 minutes. There was also no difference with respect to estimated blood loss (A vs. B, 173 ± 15 vs. 210 ± 27 ccs) and no donor in either group required intra- or post-operative blood transfusion. Total hospital analgesia in morphine sulfate equivalents for the minimal posterior supracostal approach was significantly less when compared to the conventional open approach (A vs. B, 10.7 ± 2.4 vs. 13.0 ± 2.3 morphine sulfate equivalent:1=10 mg of morphine sulfate; p <0.001). The length of hospital stay was significantly shorter for the minimally invasive supracostal group (A vs. B, 3.3 ± 0.2 vs. 4.5± 0.2 days; p<0.001).

The donor incision was measured at the first clinic visit approximately one month following surgery. Patients with the minimally invasive supracostal incision had a shorter incision (A vs. B, 16.1 ± 0.5 vs. 25.7 ± 1.5 cm; p<0.001) when compared to the conventional open approach. Longer incisions were used in the early experience with the supracostal approach. For the last ten patients the average length has decreased to 12.0 ± 0.7 cm. Finally, return to activity was significantly enhanced by the supracostal approach (Group A vs. Group B 3.9 ± 1.1 vs. 8.2 ± 1.5 weeks; p<0.001).

No recipient in either group experienced delayed graft function. The recipient's serum creatinine at three months post-transplant was equivalent (A vs. B, 1.4 ± 0.7 vs. 1.53 ± 0.1 mg/dl).

Discussion

Conventional open extraperitoneal donor nephrectomy is widely accepted and has proven to be safe and effective. Many centers have reported large series without mortality and with minimal morbidity (16-19). In this study, we demonstrate that the minimally invasive posterior supracostal approach is superior to conventional open donor nephrectomy and allows a significantly shorter incision length, minimizes postoperative analgesia, reduces hospital stay, and enhances return to full activity. In general, an excellent kidney is produced, which functions immediately, and the recipient does not experience delayed graft function.

One of the criticisms of conventional open donor extraperitoneal nephrectomy has been the surgical morbidity for the donor (20,21). The long incision of the conventional open approach can be painful and can impair a quick convalescence. For many donors a lengthy recovery period is financially unfeasible. For others, a prolonged period of physical restriction following surgery is unacceptable. In addition, our informal survey showed that younger donors find the anterior or ventral extension of the incision to be cosmetically unappealing. Our current study shows that, with the supracostal approach, a consistently shorter incision is obtained. In experienced hands, incision length less than 12 cm has become common.

The length of stay was significantly shorter when compared to historical conventional open donor nephrectomy. However, because this was a retrospective study, the results regarding hospital length of stay are difficult to interpret, as the length of hospitalization has decreased for most diagnostic related groups in the past few years. More recently, many patients undergoing surgery with the minimally invasive posterior supracostal approach have been discharged on the second postoperative hospital day. We do not believe a discharge this early would be routinely possible using conventional open door nephrectomy.

In Turner Warwick's original description the protective effect of the rib against herniation is pointed out to be a potential benefit. With an average follow up of over a year no patient has returned with a fascial defect suggesting a hernia. The use of this incision also avoids the uncommon but recognized complication of a bulge on the ipsilateral side that can occur with conventional donor nephrectomy (14). This complication is alleged to be secondary to a denervated intercostal muscular segment. Since the nerve is never in jeopardy with the use of this approach, this complication has not been seen with the modified Turner-Warwick incision. Cosmetically, younger donors have favorably received the short supracostal incision (our informal survey). The scar is not visible at all from the frontal view.

One the of the benefits of recently popularized laparoscopic donor nephrectomy has been that the return to normal activity as defined by the ability to resume normal housework, drive an automobile or return to preoperative employment is significantly shorter when compared to conventional open donor nephrectomy (22, 23). In our current study several outcome parameters including the length of stay and return to activity appear to be comparable to that of the published series of laparoscopic donor nephrectomy (19, 20, 24) (table 2). Although, controversy has surrounded the procedure regarding donor safety and graft survival (25), the novelty of laparoscopic live donor operation has prompted many transplant centers to adopt the procedure (20,21,24,26). Besides several observed complications including open conversion, reoperation, and bowel injury in the donors (23), the clinical compromise with laparoscopic donor nephrectomy appears to be the recovery of a less optimal donor kidney with reports of graft loss secondary to technical complications, delayed graft function, and even ureteral injuries requiring late reconstruction (25). It is clear that a steep learning curve exists as with any new surgical procedure (23, 26, 27) and some surgeons have appropriately questioned whether the cost of improved donor benefits are transferred to the recipient with poor allograft function and increase rate of mortality (25). The relative contraindication for the laparoscopic donor nephrectomy is a history of extensive upper abdominal surgery, which has resulted in dense adhesions (23). In addition laparoscopic donor nephrectomy is more difficult in obese patients and until recently, the procedure restricted to the right kidney (19-21, 23- 25). The immediate function of the donor kidney using the supracostal approach is not compromised, as reflected by the absence of delayed graft function postoperatively and a normal discharge creatinine in the recipient. In contrast to laparoscopic donor nephrectomy, the minimally invasive posterior supracostal approach is retro-peritoneal and not limited to thin patients. While, difficult vascular dissection and potential loss of control must be handled with immediate extension of the incision, the kidney is easily accessible even in obese donors without extending the length of the incision. Unlike laparoscopic nephrectomy, the choice of the right or left kidney is not overbearing limitation of the procedure. Either kidney can be easily taken using the posterior supracostal approach. Because supracostal approach is a modification of conventional open donor nephrectomy, it should not require special training and use of costly laparoscopic instrumentations; the surgeons performing conventional open donor nephrectomy can easily adopt the procedure.

In summary, the modified Turner Warwick approach allows a significantly shorter incision and reduces hospital stay when compared to conventional open donor nephrectomy. With this approach, the post-operative analgesic requirement is significantly less, and patients can return sooner to their full activity. The posterior supracostal approach produces a kidney which functions immediately and is associated with graft survival and function comparable to that obtained using the conventional technique.