Retroperitoneoscopic Hand-Assisted Nephroureterectomy Using a Homemade Device

Introduction

Laparoscopic nephroureterectomy (pure or hand-assisted) with excision of a bladder cuff is a minimally invasive surgical option for the treatment of upper tract urothelial cancer, with the benefit of early patient recovery. An incision for a hand-assisted procedure is a rational option when nephroureterectomy is carried out for pathologic or functional (for subsequent open surgery) purposes. This procedure is easier than pure laparoscopic surgery, while offering a comparable short-term recovery. The long-term cancer-specific survival for patients with upper tract urothelial cancer, following hand-assisted laparoscopic surgery, is similar to that of open surgery.^1–3

We favor hand-assisted retroperitoneoscopic nephroureterectomy (HARN) for the treatment of upper tract urothelial cancer. There are many commercially available hand-assist devices for hand-assisted laparoscopic surgery. However, routine use of hand-assist devices increases surgery costs and so may restrict the routine use of hand-assisted laparoscopy. To reduce surgical costs, we devised a homemade hand-assist device for use in HARN procedures.

Patients and Methods

We performed HARN with open bladder cuff excisions for patients with upper tract urothelial cancer. If the kidney or the ureter was suspected to adhere to adjacent structures or in those with evidence of inflammation around the kidney or showing lymphadenopathy on preoperative computerized tomography, we would not consider the patient as a candidate for laparoscopic surgery, and the patient would be excluded from our study.

Patients under general anesthesia were placed in a supine, spread-eagle position with legs extended and abducted (45°–60°). ⁴ The operator stood between the abducted legs of the patient. The assistant stood at the nondominant hand side of the surgeon, beside the patient. HARN was performed through a Gibson incision and two trocars at the umbilical level (one at the midaxillary line and one just lateral to the peritoneal reflection). The incision closer to the assistant was 11 mm long for the laparoscope, and the second incision was 12 mm long for the operative instrument. After the patient was prepared and draped, a 7–8-cm Gibson incision was made on the ipsilateral side of the lesion. Working through the Gibson incision, peritoneal and retroperitoneal fat was then bluntly detached from the abdominal wall until a retroperitoneal space was created underlying the first trocar site at the umbilical level of the midaxillary line. The first trocar was inserted under hand guidance, through the incision at the planned site in the retroperitoneal cavity. Then, a medium-sized plastic wound retractor (Alexis; Applied Medical, Rancho Santa Margarita, CA) was placed in position through the incision, with the bottom ring (green) inside the abdomen and external ring (white) outside of the abdomen (Fig. 1A). The external ring of the Alexis wound retractor was folded to minimize the gap between the bottom and external rings. When double-gloving, the sizing of gloves is dependent on individual needs. The outermost glove is generally a half a size larger than the inner glove. The surgeon inserted the double-gloved, nondominant hand into the retroperitoneal space through the wound retractor space (Fig. 1B). The cuff of the outer surgical glove was then turned outside in and snapped onto the external ring of the wound retractor (white) (Fig. 1C). Pneumoretroperitoneum was created at 15 mm Hg carbon dioxide; under this pressure, the peritoneum spontaneously detached from the abdominal wall, and a retroperitoneal space was created for laparoscopic surgery. A 10-mm 0° laparoscope was inserted through the first trocar, and the peritoneal reflection was identified. The second trocar was then inserted into the retroperitoneal space under hand and laparoscope guidance, followed by insertion of the laparoscope and operation instrument into separate trocars (Fig. 1D).

OPEN IN VIEWER

FIG. 1.

(A) A plastic wound retractor is placed in position through the Gibson incision after insertion of the first trocar. (B) The surgeon's double-gloved, nondominant hand is inserted into the retroperitoneal space through the retracted wound. (C) The cuff of the outer sterile surgical glove is turned outside in and snapped onto the external ring of the wound retractor. (D) Carbon dioxide pneumoretroperitoneum is created at 15 mm Hg. The second trocar is inserted into the retroperitoneal space with hand and laparoscope guidance.

After the kidney was palpated, the posterior part of Gerota's fascia was dissected to the posterior aspect of the kidney. The ureter was traced from the lower pole of the kidney to the level of the external iliac vessels and to the renal hilum. The ureter was next traced upward to the renal pedicle. The renal vein and artery were identified, clamped, and transected. Pulling the kidney downward facilitated dissection of the remaining attachment of the kidney's upper pole after transection of the renal pedicle. The kidney and upper ureter were then removed though the Gibson incision after removal of the intra-abdominal hand's outer surgical glove from the external ring of the Alexis wound retractor. The retractor was kept in place to maintain retraction during the distal ureterectomy and excision of the bladder cuff. The distal ureter and bladder cuff were excised under direct vision using standard open surgical techniques. The detrusor was opened circumferentially, and a 2-cm cuff of bladder was removed with the distal ureter. The opening in the bladder was closed in two layers with 2-0 absorbable sutures.

Results

From October 2011 to March 2012, HARN and open bladder cuff excisions were performed on 24 consecutive patients with the aid of a homemade hand-assist device comprising an Alexis wound retractor and surgical gloves, without the use of a commercially available hand-assist device or open surgery. No patient was excluded during our study period owing to the exclusion criteria. Twenty-four patients (11 men and 13 women; age range, 35–86 years; mean, 68.3 years) were included in this study: 23 presented with gross hematuria and 1 with an incidental sonographic finding. Four patients had end-stage renal disease and were on hemodialysis. Average patient height was 163 cm (range, 153–174 cm), average weight was 68 kg (range, 53–84 kg), and average±standard deviation body mass index was 26.0±3.7 kg/m² (range, 20.1–34.1 kg/m²). Four patients met the criteria of obesity (body mass index >30 kg/m²), being 31.0, 31.8, 32.3, and 34.1 kg/m².

Mean operation time was 103 minutes (range, 88–144 minutes), and mean estimated blood loss was 81 mL (range, 50–200 mL). Mean specimen weight was 381 g (range, 250–750 g). The specimens were removed smoothly from all patients, en bloc, via identical incisions. Mean times to oral intake and ambulation were 1.5 and 2.1 days, respectively. The mean parenteral narcotic requirement (morphine) was 17.6 mg (range, 3–36 mg). All patients recovered uneventfully to normal daily activity. Mean hospital stay was 8.8 days (range, 7–13 days).

No postoperative wound complication was related to the hand-assist device. One patient developed febrile urinary tract infection within 3 weeks postoperatively and was re-admitted for parenteral antibiotics. Of the 24 patients, 13 had renal pelvis transitional cell carcinoma, 8 had ureteral transitional cell carcinoma, and 3 had renal pelvis plus ureteral transitional cell carcinoma. Pathologically, according to the TNM system, 5 lesions were Ta, 12 were T1, 4 were T2, and 3 were T3.

Discussion

To our knowledge, this is the first article describing the use of a homemade hand-assisted method, comprising a seal formed by double-gloving, connecting the outer glove to an Alexis wound retractor, and operating through a Gibson incision during retroperitoneoscopic nephroureterectomy and excision of the bladder cuff. The Alexis wound retractor provides circumferential, atraumatic retraction while protecting the wound from contamination during the entire operation; the retractor is clinically proven to decrease the risk of wound infection at the incision site. The wound retractor's tamponade effect also prevents port-side bleeding and subcutaneous emphysema. ⁵

Many different hand-assist devices currently exist. The most recent generation of device includes the LapDisc (Ethicon Endo-Surgery, Cincinnati, OH) and the GelPort (Applied Medical). Compared with the GelPort hand-assist device, our homemade device offers a dramatic cost reduction by using surgical gloves instead of a GelSeal cap to prevent air leakage and maintain the pneumoretroperitoneum during HARN. Lee et al. ⁶ also used a surgical glove to cover the outer ring of a small-sized wound retractor, preventing carbon dioxide gas leakage and serving as a multichannel port during the performance of laparoscopic nephroureterectomy. The air seal and maintenance of pneumoretroperitoneum are continuously effective while the surgical glove cuff covers the external ring of the wound retractor. The initial setup of our homemade hand-assist device is simple and quick and does not require lubricating jelly for its insertion, or to enable a back-and-forth movement of the hand, as the GelPort device requires.

Surgeons may express concern about interference in the progress of hand-assisted surgery due to restricted dexterity and loss of tactile sensitivity resulting from double-gloving. In the study of Fry et al., ⁷ double-gloving did not have a substantial impact on manual dexterity or tactile sensitivity compared with single-gloving. We adopted the practice of doubling-gloving since introduction of hand-assisted laparoscopic surgery. Double-gloving is an effective method to reduce surgeons' potential for contact with bodily fluids and exposure to bloodborne pathogens. ⁸

Another concern is that the range of intra-abdominal hand motion could be limited, particularly when the surgeon needs to move his or her hand further during dissection. Surgical gloves provide high strength and elasticity, and thus the intra-abdominal hand can reach the upper pole of the kidney during dissection with a broad range of hand movement, without discomfort or compromise. However, rigorous hand movements, or encountering adhesion during upper kidney dissection, may cause the outer glove cuff to slip or tear from the external ring of the wound retractor and cause a carbon dioxide gas leak. In our experience, use of an outer surgical glove with a longer cuff solves this problem.

A further drawback is that our homemade hand-assist device could not maintain pneumoretroperitoneum during the hand exchange. All commercially available hand ports, with the exception of GelPort, can maintain the pneumoretroperitoneum during hand exchange. Our homemade hand-assist device is easy to set up and easily restores pneumoretroperitoneum following the hand exchange.

Hand-assisted laparoscopy is associated with neuromuscular strain to the surgeon's upper extremity. The commercially available LapDisc incorporates a silicone membrane attached to three rings that tighten around the hand to create a seal, whereas the GelPort device uses a gel cap to create and maintain a seal around the inserted hand. The maximum forearm compression forces for the LapDisc is 97 mm Hg, and that for the GelPort is 78 mm Hg. ⁹ Implications of forearm compression include surgeon fatigue and temporary, or permanent, neurologic or vascular deficits. ¹⁰ Unlike these commercial hand-assist devices, which function by compressing the forearm of the intra-abdominal hand to prevent air leakage, the air seal effect of our homemade hand-assist device is provided by a surgical glove that covers the external ring of the wound retractor. It appears to exert less compressive force on the surgeon's forearm than either the LapDisc or GelPort does. However, this observation needs further study for confirmation.

The mean operation time for HARN and bladder cuff excision using our homemade hand-assist device was 103 minutes (range, 88–144 minutes), less than that described in our previous report of HARN and open bladder cuff resection using the GelPort hand-assist device (average, 137.6 minutes; range, 85–220 minutes). ⁴ This improvement was probably due to our increased familiarity with hand-assisted surgery. Our experience demonstrates that the homemade hand-assist device does not increase the difficulty of HARN surgery and does not slow the operation. Our homemade device offers a dramatic cost reduction by using an Alexis wound retractor ($75) and a pair of surgical gloves ($1) instead of a GelPort ($728) as the hand-assist device to prevent air leakage and maintain the pneumoretroperitoneum during HARN at our institution. Our long hospitalization is due to the Taiwan health insurance policy. Insurance premiums are very low for Taiwanese citizens covered by the National Health Insurance. Patients were not discharged until they felt they had recovered completely.

Conclusions

Our homemade hand-assist device, comprising an Alexis wound retractor and surgical gloves, is both feasible and safe in HARN. It is more cost-effective than commercially available hand-assist devices. It offers ease of use and is efficacious at maintaining insufflation. Our device provides an adequate range of hand motion and appears to exert minimal compressive force on the surgeon's forearm.