Thick vs thin loop transurethral resection of the prostate: a double-blind prospective trial of early morbidity

Authors


Abstract

Objective To compare blood loss, irrigation requirements and hospital stay between standard transurethral resection of the prostate (TURP) and resection incorporating vaporization, i.e. transurethral vaporization resection (TUVRP).

Patients and methods Seventy patients were prospectively randomized in a blinded fashion between TURP with a standard resection loop (thin loop) or resection with a roller cutting electrode (thick loop). The outcome was assessed as the change in haemoglobin at 30 min and 24 h after the resection finished. Secondary outcome measures were irrigation requirements, length of catheterization, hospital stay, changes in serum sodium, and complications. The nursing staff and surgical registrar managing the patients after surgery were unaware of the resection technique used.

Results Data were available for 65 patients (36 thin loop and 29 thick loop resections). The mean decrease in haemoglobin after TURP was 14.8 g/L at 30 min and 11.8 g/L at 24 h; for TUVRP the corresponding values were 14.6 and 14.2 g/L (P = 0.69). In addition, there was no difference in irrigation requirements, changes in serum sodium, length of catheterization, hospital stay or bleeding complications.

Conclusion The use of a thick loop at higher cutting current settings during TUVRP offers no advantage over standard thin-loop TURP.

Introduction

TURP remains the main surgical treatment for men with LUTS considered to be secondary to obstruction by the prostate [1]. Alternative surgical treatment options have been explored in an attempt to reduce the risk of blood transfusion (2.5–4.2%) [2,3], clot retention (5.5%) [1] and TURP syndrome (2–3%) [2–4].

Recent surgical innovations include transurethral electrovaporization of the prostate (TUVP) [5–7] and holmium laser resection of the prostate [8,9]. Both techniques had equivalent efficacy to TURP in prospective randomized trials, but with reduced blood loss [5–7,9] and glycine absorption [9,10].

While holmium laser resection of the prostate requires an expensive holmium:YAG laser, TUVP uses existing diathermy equipment and is therefore an attractive option. However, TUVP is associated with dysuria [11] and increased irritative voiding symptoms in the first 6 weeks after resection [7]. TUVP is theoretically less effective for larger prostates [1]. Shokier et al.[10] limited their trial to prostates of leqslant R: less-than-or-eq, slant 60 g, as did Kaplan et al.[5], while Michel et al.[11] noted a 70% loss of efficiency in tissue ablation with repeated movement of the roller-ball in the same area.

However, it is theoretically possible to combine tissue resection with tissue coagulation by the use of a ‘thick’ or band loop. Thick loops are about four times wider than ordinary thin (0.3 mm) loops but have the same loop diameter. The increased width of the resecting element allows higher energy settings and increases the amount of energy transferred into the tissue while resecting [12]. This should produce coagulation and cutting, as opposed to the almost pure cut achieved with a standard resection loop. There are claims that this results in less bleeding at the time of TURP [12,13]. This new approach has been termed transurethral vaporized resection of the prostate (TUVRP).

To effectively evaluate this new technology, we established a prospective blinded trial in 1998 to assess the early morbidity of thick-loop and standard thin-loop resection. The primary outcome of interest was the change in haemoglobin and serum sodium levels over the first 24 h after surgery. Secondary outcome measures were indirect measures of blood loss at or after resection. These were taken as the amount of irrigant used during and after resection, and the length of time a catheter was present after surgery. A single multicentre randomized trial of TURP vs ‘band-loop’ resection was published recently [14].

Patients and methods

Between June 1998 and November 1999, after obtaining ethical approval, a prospective blinded randomized trial was conducted comparing standard TURP using a thin loop against resection with a thick loop (a roller cutting electrode, Karl Storz GmbH, Germany). Patients were selected for the trial if they had been placed on a waiting list for TURP either on clinical (urinary retention or a maximum urinary flow rate of < 15 mL/s) or symptomatic grounds (IPSS > 7). Randomization was in groups of 20, with allocation to treatment group by sealed envelopes, opened by the nursing staff during the initial cystoscopy before resection commenced.

The exclusion criteria were: previous TURP, IPSS < 7, known prostate cancer or anticoagulation with warfarin. No patient was excluded on the basis of prostate size or presence of an indwelling urethral catheter. Patients received 3 mg/kg of gentamicin intravenously 1 h before surgery if there was UTI or an indwelling catheter.

The loop used was not recorded in the operative record and the nursing staff responsible for care after surgery were unaware of the loop used, as was the registrar deciding on the timing of cessation of irrigation, catheter removal, discharge and need for a blood transfusion. Patients were evaluated for the time from the start of resection to the insertion of a catheter (operative duration), volume of irrigant used in theatre and on the ward, and the duration of catheterization and hospital stay. Serum sodium and haemoglobin levels were measured at 30 min and 24 h after resection had finished. The number of bladder washouts and the need for a blood transfusion were also recorded. Patients were reviewed at 6 weeks for the need to be re-catheterized, readmitted to hospital or for bleeding problems in the intervening period.

A standard 24 F transurethral resecting set was used with an electrosurgical generator (Force FX, Valley Laboratory, Colorado, USA). Glycine 1.5% at a height of 80 cm above the patient's symphysis pubis was used as the irrigant. For the thick loop, cutting was set at 200 W and coagulation at 80 W, while for the standard loop these settings were 120 W and 60 W, respectively. Resection with the roller cutting electrode proceeds identically to that for a standard TURP, except that the drag speed is slower, i.e. the time to resect one chip of prostate is slighter longer. In addition, because of charring the capsule is not as clearly defined and is detected by the increasing flexibility of the tissue remaining as the adenoma is resected. At the end of the procedure a 24 F three-way catheter was inserted with continuous saline irrigation.

The registrar, who was not present in theatre at the time of resection, decided the subsequent management of irrigation, catheter removal and discharge. Nursing care was standard for both groups.

To investigate any differences in the change in haemoglobin and sodium level between the treatment groups a repeated-measures mixed model was used, including the covariates of seniority of surgeon, aspirin use, presence of a catheter, age, weight of tissue resected and operative duration. The outcome of interest was the time × treatment loop interaction. The effect of different treatments on the amounts of saline irrigation, glycine irrigation and length of catheterization was investigated using anova, including the same variables as above. Log-transformed data were used for saline irrigation and length of catheterization. The length of stay in hospital was converted into binary data by categorising it as < or > 2 days; a logistic regression analysis was used with the same covariates on this binary data.

Results

In all, 70 patients were enrolled into the study; five were unavailable for analysis, three had no documentation available after randomization, one had had previous surgery for prostate cancer, and one had a urethral stricture. Both the last two patients had been randomized to the standard loop. The subsequent analysis is therefore based on 65 patients (36 standard and 29 thick-loop resections; mean age 71 years, range 51–87). Of these, 22 patients were on aspirin and 18 were catheterized. The weight of tissue resected was 3–107 g. Most patients were operated on by either the Registrar or Fellow, under consultant supervision. The baseline variables for the two treatment arms are given in Table 1.

Table 1.  Baseline characteristics and perioperative variables for the two treatment groups
Baseline variablesThick loopThin loop
  1. * Median (10–90th percentile); †  anova; ‡ logistic regression analysis on covariates of patient age, aspirin, presence of catheter before surgery, operative duration, weight of resected tissue and seniority of surgeon.

Number2936
Age (years)*68 (60–80)71.5 (60–79)
No. on aspirin (%)12 (41)10 (27)
No. catheterized before (%)7 (24)11 (31)
Confounding variables*
Operative duration (min)30 (18–41)29 (14–55)
Eeight of resected tissue (g)13 (4–38)11.5 (4–39)
No. of operations21 (72)28 (78)
 by registrar or fellow (%)
Outcome variables
Glycine irrigation (L)*12 (8–18)12 (6–22)
 F1,56 = 1.08, P = 0.3
Saline irrigation (L)*9 (7–67)8 (1–46)
 F1,54 = 0.61, P = 0.44
Length of catheterization (h)*20 (14–64)21 (1–46)
 F1,54 = 0.01, P = 0.91
Hospital stay (days)
 < 21118
X21 = 2.2, P = 0.14
 ≥ 21818

The volume of irrigation with glycine and saline, and duration of catheterization and hospital stay for each treatment arm are also given in Table 1. There was no evidence of any difference between treatments in the outcome variables (Table 1).

After TURP with a standard loop the mean decrease in haemoglobin was 14.8 g/L at 30 min and 11.8 g/L at 24 h; for thick-loop resection these values were 14.6 and 14.2 g/L, respectively (Fig. 1). There was no indication of a difference in the change of haemoglobin in the two groups (F2,115 = 0.37, P = 0.69).

Figure 1.

The changes in haemoglobin after resection. The horizontal line is the median, the box the 25–75th percentile and the bars (whiskers) the lesser of either the sample range or 1.5× the interquartile range. The circles represent the outlier values, the asterisks the extreme values. Thin loop, green and light green; thick loop, red and light red.

For serum sodium the decreases were 0.5 mmol/L at 30 min and 3.2 mmol/L at 24 h for the standard loop, and 1.7 and 2.6 mmol/L, respectively, for the thick loop. There was no evidence of a difference in the change in sodium in the two groups over time (F2,115 = 0.55, P = 0.58).

Thirteen patients were irrigated for clots after surgery (seven thick loop and six standard loop). Adverse events occurring after surgery are given in Table 2. One patient received a blood transfusion of two units (thick-loop arm). Eleven patients had some problem related to haematuria after their first trial of voiding. As some patients had more than one problem, the totals in each column of Table 2 are not arithmetic sums. Using Fisher's exact test there was no difference between the treatment arms in the total number of patients who had problems after surgery (P = 0.65).

Table 2.  Adverse events in each group
Adverse eventThick loopThin loop
  • * 

    includes catheterized for haematuria.

Re-catheterized for bleeding
 while an inpatient
32
Transfused1 (2 units)
Bleeding problems after discharge,
 not readmitted*
23
Readmitted with bleeding problems*22
Total n (%) of patients with a problem5 (17)6 (17)

Discussion

TUVRP appears to be a logical progression from the pure vaporization as used in TUVP and offers the potential to resect the prostate with reduced blood loss but with no requirement for expensive new equipment. Vaporization of the prostate with, e.g. the Vaportrode® electrode, is associated with reduced blood loss both during and after surgery, but with similar clinical and objective outcomes as those after TURP [5–7,10]. Reservations have been expressed about the initial irritative symptoms produced [7,11], although this was not noted by either Kaplan et al.[5] or Shokeir et al.[10]. In addition, there would appear to be some limitations on prostate size using vaporization alone [1,5,10]. To overcome these drawbacks thickened loops were developed, which theoretically should produce coagulation while resecting [12]. A 2-mm coagulation zone at the chip margin, not seen with TURP, has been shown with TUVRP in a dog model, suggesting that thick loops should be more haemostatic than standard resection [13]. Initial open single-arm trials of different thick loops claimed that general ooze and blood loss was less than would be expected with TURP [12,13,15].

Perlmütter and Vallancien [16] reported on 91 patients who underwent resection with the VaporCut® loop and 65 patients who had a ‘wedge’ resection, with a 1-year follow-up. The cutting current was set at 300 W for the wedge and 120–150 W for VaporCut resection. The only apparent benefit claimed over TURP was reduced blood loss during resection, enabling a safer operation. The patient's peak urinary flow rates and IPSS after resection with each device were comparable to those after standard thin-loop TURP. Perlmütter and Schulsinger [13] previously reported on the 65 patients who received a wedge resection and noted that better coagulation was obtained at 275–300 W as opposed to 200–250 W.

Talic [15] described 31 patients who underwent vaporized resection with the ‘wing’ electrode, 19 of whom were in urinary retention. The mean decrease in haemoglobin was 25 mL/L at 1 h and 34 mL/L at 24 h. The changes in serum sodium level were similar to those in the present study. The mean weight of resected tissue, at 28 g, was greater than that in the present study. A dramatic reduction in general ooze was reported.

In the present double-blind prospective trial there was no evidence of any such reduction in blood loss; using a mixed linear model there was no statistically significant difference between standard TURP and one incorporating a thick loop. As can be readily appreciated from Fig. 1, there is no clinical difference between the treatment arms. In addition, there was no evidence of any significant difference in clinically useful outcome measures, e.g. the amount of irrigant used, length of catheterization, complications ascribed to bleeding or even length of hospital stay. Finally, there was no safety benefit, in that serum sodium changes were the same in both treatment arms.

Two senior registrars under consultant supervision carried out most of the resections, which may represent a weakness of this study; however, this can also be viewed as a strength, as the percentage of procedures carried out by a registrar was similar in the two treatment arms, as were the other potential confounders (aspirin, indwelling urethral catheter and weight of prostate resected). The current used for resection in the present TUVRP arm (200 W) was lower than that use in other published studies. One consultant (P.B.) and the fellow (M.H.) had used TUVRP for ≈ 1 year before initiating this trial. During one of these vaporized resections electrical arcing at 240 W destroyed a resectoscope lens. This issue was discussed by Perlmütter and Vallancien [16], although it is not clear if they had a similar problem. Hence a lower current was selected for the present trial.

Gotoh et al.[14] randomized 53 patients in a multicentre trial to either band-loop or standard loop TURP; the former was conducted at 230–250 W. There was no difference in the duration of catheterization or time to disappearance of macroscopic haematuria after resection. In addition, the maximum flow rate, changes in symptom scores and postvoid residual volume did not differ between the treatment arms. The present single-institution study therefore confirms and extends the findings of the only other randomized trial of thin-loop or thick-loop resection [14]. There was no prostate size restriction on patients in the present study, whereas there was in the study of Gotoh et al.[14], and the primary outcome in the present study (serial haemoglobin measurements) was the same in each treatment arm. These findings do not appear to be specific to loop type. Gotoh et al.[14] used a band-loop while we used a roller-cutting electrode. Although the design of the band-loop and roller cutting electrode differ, both are designed as thick loops to incorporate vaporization during resection. In conclusion, thick-loop vaporized resection of the prostate at 200 W cutting current offers no significant haemostatic or clinical benefits over standard TURP.

Acknowledgements

We thank the nursing staff of theatre 5 and ward 8 at Waikato hospital, without whom this study would not have been possible.

M. Holmes, Urology Services Limited, PO Box 19210, Hamilton, New Zealand.
e-mail: UROLOGY@hwl.co.nz

Ancillary