Prospective non-randomized evaluation of four mediators of the systemic response after extraperitoneal laparoscopic and open retropubic radical prostatectomy

Authors


Andreas Jurczok, Department of Urology, Medical Faculty, Martin Luther University, Halle-Wittenberg, Ernst-Grube-Str. 40 06120 Halle/Saale, Germany.
e-mail: andreas.jurczok@medizin.uni-halle.de

Abstract

Associate Editor

Ash Tewari

Editorial Board

Ralph Clayman, USA

Inderbir Gill, USA

Roger Kirby, UK

Mani Menon, USA

OBJECTIVE

To report a prospective, controlled, non-randomized patient study to determine the systemic response to extraperitoneal laparoscopic (eLRP) and open retropubic radical prostatectomy (RRP).

PATIENTS AND METHODS

In all, 403 patients who had eLRP (163) or open RRP (240) were recruited; patients in both groups had similar preoperative staging. In addition to peri-operative variables (operative duration, complications, blood loss, transfusion rate, hospitalization, catheterization), oncological data (Gleason score, pathological stage, positive margins) were also compared. The extent of the systemic response to surgery-induced tissue trauma was measured in all patients, by assessing the levels of acute-phase markers C-reactive protein (CRP), serum amyloid A (SAA), interleukin-6 (IL-6) and IL-10 before, during and after RP.

RESULTS

The duration of surgery, transfusion rate, hospital stay and duration of catheterization were comparable with those in previous studies. There was an increase in IL-6, CRP and SAA but no change in IL-10, and no differences between eLRP and RRP over the entire period assessed.

CONCLUSION

The invasiveness of eLRP could not be substantiated objectively based on the variables measured in this study. The surgical trauma and associated invasiveness of both methods were equivalent.

Abbreviations
(e)(L)(R)RP

(extraperitoneal) (laparoscopic) (retropubic) radical prostatectomy

APR

acute-phase response

IL

interleukin

CRP

C-reactive protein

SAA

serum amyloid A.

INTRODUCTION

During the past two decades radical retropubic prostatectomy (RRP) has been continually refined. In addition, the introduction of the nerve-sparing technique to preserve erectile response has further improved the quality of RP [1]. Following these advances, the development of laparoscopic RP (LRP) in the late 1990s resulted in further innovation and motivation [2,3]. Regardless of the ongoing analysis of functional and oncological results, the use of LRP, initially introduced using a transperitoneal approach, and later with extraperitoneal technical variants (eLRP) has increased, particularly in France and Germany [4,5].

However, initial prospective studies comparing quality of life and convalescence showed no clear advantage of LRP over conventional RP [6]. Currently, data available on morbidity, e.g. intraoperative blood loss, incidence of complications, postoperative requirement of analgesics and duration of hospitalization, show some discrepancy and do not allow a definitive conclusion. Clearly, laparoscopic procedures result in less blood loss than open procedures, whereas oncological and functional variables appear to be comparable [2,3].

To date, no study has reproducibly examined the extent of surgical trauma and the subsequent systemic response of both operative techniques using quantifiable variables. Tissue damage from an operation causes a local activation of various cells (e.g. monocytes and macrophages), which release cytokines and other mediators [7,8]. This activation is either limited to a local level or can be followed by a systemic acute-phase reaction. Experimental and clinical data suggest that the acute-phase response (APR) to tissue damage reflects a benefit for the patient, and that the intensity of an APR is somewhat proportional to the extent of tissue damage [9–11]. However, tissue damage from surgery can also result in postoperative fatigue and effects on various organ systems, with possible organ failure [10]. A systemic reaction from an accidental trauma might be helpful in stabilizing vital functions, whereas in elective surgery under controlled conditions the systemic reaction should be as low as possible.

In the present prospective, controlled, non-randomized study we attempted to quantify the surgical trauma of both conventional RRP and eLRP. To manage the complexity of the APR, the pro-inflammatory cytokine interleukin (IL)-6 and the modulatory cytokine IL-10 were examined [7,8,12], as were C-reactive protein (CRP) and serum amyloid A (SAA), both important acute-phase proteins [7,8,13].

PATIENTS AND METHODS

Between January 2003 and April 2006, 403 patients were included in the present study; all had RP for a clinical locally confined prostate carcinoma that had been confirmed histologically (Table 1). The distribution of patients in each treatment group was not randomized, but inclusion criteria for each of the treatment groups were equal. There were no significant differences between the treatment groups in preoperative variables (Table 1).

Table 1.  The patients’ characteristics before, during and after eLRP and RRP
VariableeLRPRRPP*
  • *

    No value means the difference was not significant.

Number of patients163240 
Median (range):
 Age, years 62.9 (42–74) 64.8 (52–76) 
 Preoperative PSA level, ng/mL  7.9 (2.4–10.2)  7.25 (4.4–11.3) 
 Gleason score (biopsy)  5.7  5.3 
Preoperative clinical stage, %
 T1a  0  6 
 T1c 79 75 
 T2a 14 12 
 T2b  7  7 
Median (range):
 Operative duration, min180 (120–240)120 (80–190)<0.05
 Blood loss, mL200 (100–700)550 (200–1900)<0.05
Transfusion, %  3%  9%<0.05
Complications, %
 rectal lesion  1.8  1.6 
 lymphocele  3.2  2.9 
 wound infection  3.1  3.4 
 revision  1.25  2.5 
Duration of catheterization, days  8.9 10.2<0.05
Analgesic requirement, (mg morphine equivalent) 33 35 
Hospitalization, days  9.4  11.2 
Histopathology
Mean (range) prostate weight, g 37 (18–72) 42.3 (20–120) 
Median Gleason score (specimen)  6.4  5.7 
Pathological stage, %
 PT2a 16 19 
 PT2b 27 22 
 PT2c 23 25 
 PT3a/b 33 34 
Positive lymph nodes, %  0  1.67 
Positive surgical margins, %
 pT2a/b/c  9.8 12.6 
 pT3a/b 29 31 

In all, 240 patients had RRP (ascending) with pelvic lymph node dissection, as detailed by Walsh [14]. Each surgical procedure began with a pelvic lymphadenectomy. The RRP procedure is well established in our institution and was performed by three experienced surgeons. There was no difference between the operative results of each.

For eLRP, surgery proceeded according to the pre-peritoneal technique published by several authors [15]. A pelvic lymphadenectomy was performed in all cases of eLRP; the eLRP procedure is well established in our institution and was performed by three experienced surgeons, with no difference between the operative results of each.

Clinical variables included the duration of surgery, intraoperative blood loss, blood transfusion rate, complication rate, duration of hospitalization and duration of urinary catheterization. For oncological cases, the variables evaluated were Gleason score, pathological stage and surgical margin status.

Blood samples were collected before, during and after surgery to determine serum levels of CRP, SAA, IL-6 and IL-10, at times t0 (24 h before RP) and then t1 (after placing trocars/after incision of the skin), t2 (after completing the urethrovesical anastomosis), and then at t3–t8 (6, 12, 24, 36, 48 and 72 h after RP). In each case, 7.5 mL of blood was aspirated; serum was separated by centrifugation, aliquoted and stored at −70 °C until required. As cytokines are generally unstable molecules, all specimens were collected and handled carefully using standardized conditions. Serum levels of IL-6 and IL-10 were measured using chemiluminescent immunometric tests, automatically assayed on the Immulite analyser (Diagnostic Products, Los Angeles, CA, USA). SAA and CRP were measured by turbidimetry (Behringwerke, Marburg, Germany). All tests were performed according to the manufacturers’ instructions.

The data are reported as the mean (sd) and were compared using the Student’s two-tailed t-test. Comparisons were considered statistically significant with P < 0.05. For statistical analyses, the results of serum variables were calculated as the median (5%, 25%, 75% and 95% percentile). For comparison of paired values within the same group, the Friedman test (nonparametric) was used, followed by the Wilcoxon-Wilcox test. For comparison of unpaired values among different groups, the nonparametric Kruskal–Wallis test (one-way anova by ranks) was used, followed by the Tukey–Kramer test; again P < 0.05 was considered to indicate statistical significance. Based on a power of 0.8 and a significance level of P < 0.05, 50 patients in each group was recommended. The estimated difference between the groups was 10% and the α error was 5%.

RESULTS

The median duration of surgery was 180 (120–240) min for eLRP and 120  (80–190) min for RRP (P < 0.05). Intraoperative blood loss was estimated for patients in the eLRP group, as the bladder is opened early when the descending technique is used; the mean loss was 200 (100–700) and 550 (200–1900) mL in the eLRP and RRP groups, respectively (P < 0.05). The transfusion rate was 3% in the eLRP and 9% in the RRP group (P < 0.05).

Three patients in the eLRP group (1.8%) and four in the RRP group (1.6%) presented with a rectal lesion, which required intraoperative suturing. However, no conversion was necessary in the eLRP group. The duration of catheterization was 8.9 days in the eLRP group and 10.2 days in the RRP group (P < 0.05). There was no difference in analgesic usage after RP between the groups; the hospitalization in the eLRP group seemed to be shorter than that in the RRP group, but there was no significant difference (9.4 vs 12.2 days, P > 0.05). The peri-operative data are summarized in Table 1.

There was no difference in Gleason score after RP (5.7 vs 6.4) and pathological stages between the eLRP and RRP groups. The surgical margins were positive in 9.8% of patients with T2 tumours in the eLRP group, and in 12.6% of patients with T2 tumours in the RRP group (P > 0.05). Four cases of positive lymph nodes were detected in the RRP group (1.6%); the oncological variables are also shown in Table 1.

There was an increase in the level of CRP detected 12 h after surgery, with a further increase of >100 mg/L at 48 h; there were no differences in serum CRP levels between the treatment groups (Fig. 1A). There was no relevant increase in SAA levels until 6 h after surgery (t3, Fig. 1B), but at 12–48 h (t5–t7, Fig. 1B) there was a clear increase in the level of SAA (t5–t7) in both treatment groups (P < 0.05; Fig. 1b) but no difference between the groups in SAA levels. There were no increases in initial IL-6 levels at t0 and t1; after urethrovesical anastomosis (t2) there was an increase in IL-6 in both groups (P < 0.05) but no difference in IL-6 levels between the groups (Fig. 1C). The eLRP and RRP groups had similar serum concentrations of IL-10 over the entire observation period (Fig. 1D).

Figure 1.

Serum concentrations of: A, CRP; B, SAA; C, IL-6; and D, IL-10 for eLRP (163 men) and RRP (240 men), shown as the median (5%, 25%, 75% and 95% percentiles). There were no differences between the treatment groups (P > 0.05, Kruskal–Wallis test), but significant increases in: A, CRP within the treatment groups after sampling time t3 (t0–t3 vs t4–t8; P < 0.05); B, SAA within the treatment groups after sampling time t4 (t0–t4 vs t5–t8; P < 0.05); and C, IL-6 within the treatment groups after sampling time t1 (t0–t1 vs t2–t8; P < 0.05; all Friedmann test, Wilcoxon-Wilcox test).

DISCUSSION

Although eLRP is a feasible procedure, its superiority over open surgery has not yet been confirmed. The benefits of a shorter hospital stay, reduced postoperative pain, and shorter convalescence have been reported in comparisons between patients undergoing laparoscopic and open nephrectomy, nephroureterectomy, and adrenalectomy [16–18].

In the present study, all peri-operative data (duration of surgery, blood loss and duration of catheterization) were comparable with previous studies. Blood loss during the retropubic procedure is highly variable [19] but with improved control of the plexus of Santorini, blood loss can be reduced to <400 mL [1]. The most reproducible way of estimating blood loss is by the proportion of transfused patients. The transfusion rate for RRP is reportedly 0.5–29%[1]; the proportion of patients transfused after laparoscopic surgery is relatively low (2.8–3%) [20]. In the present study the transfusion rate associated with eLRP was significantly lower than for RRP (3% vs 9%) with a mean blood loss of 200 vs 550 mL. Although there is a significant difference between the techniques in blood loss and transfusion rate, there was no significant difference in the level of the serum markers. However, it is well known that significant blood loss has a traumatic impact on the patient, which can be measured by cytokines [21]. There appears to be an absolute threshold for triggering an APR under which a trauma causes no significant systemic response [22]. This threshold was obviously not reached, as shown by the different blood loss of both groups in the study.

According to current reports there are no differences in the oncological outcome of patients treated by RRP or LRP [2,3]. The surgical margins were positive in 9.8% of patients with T2 tumours in the eLRP group, and in 12.6% of patients with T2 tumours in the RRP group, with no significant difference.

The duration of catheterization was 8.9 days in the eLRP group and 10.2 days in the RRP group (P < 0.05); creating the urethrovesical anastomosis under visual control during laparoscopic surgery allows the catheter to be withdrawn early [23]. The bladder catheter, which limits physical activity and is one of the most troublesome aspects for the patient [24], can therefore be withdrawn earlier than 14–21 days, the period usually considered necessary for healing of the urinary epithelium. This observation has led surgeons who use open surgery to adopt a similar approach [25].

There was no difference in the use of analgesics after RP between the treatment groups; the hospitalization of patients in the eLRP group appeared to be shorter than in the RRP group, but not significantly. The duration of hospitalization represents an important part of the cost of RP and is influenced by several factors, e.g. age, resumption of feeding, use of peridural anaesthesia, postoperative analgesia and the health system [26]. In Germany, following the introduction of the DRG-system in 2006, the duration of hospitalization was significantly shortened [27].

In the present study, three patients in the eLRP group (1.8%) and four in the RRP group (1.6%) had a rectal lesion, which required intraoperative suturing. In the eLRP group, no conversion was necessary. The reported frequency of rectal and ureteric injury in eLRP is similar to that for open surgical procedures [28]. Rectal wounds do not always require conversion to open surgery, but can be repaired laparoscopically [28].

The aim of this study was to determine the systemic response to eLRP and open RRP; the impact of surgically caused tissue damage was reported in several experimental and clinical studies [9–11]. There seems to be an absolute threshold for triggering an APR under which a trauma causes no significant systemic response. Therefore, the goal of minimally invasive techniques is to reduce factors that contribute to exceeding this threshold. From clinical studies we know that the laparoscopic approach provides clinical benefits for the patient, especially after surgery [29].

We investigated four different variables of the APR (IL-6, IL-10, CRP and SAA). The intraoperative timing to take blood samples depends on certain comparable sample times of both surgical techniques, e.g. placing the trocars/incision of the skin, and completing the urethrovesical anastomosis. This is an empirical timetable, in particular during RP, but significant changes in serum levels would be recognized in both groups.

The APR is initiated and controlled by many cells and mediators; these mediators represent the elements of communication between vessels, immune system, bone marrow and CNS [7,8]. Common stimuli that induce the APR include surgery, infection, chemical or physical trauma, burns, tissue infarction, and advanced cancer [7,8]. The mediators include cytokines (e.g. TNF-α), ILs, interferon-γ, growth factors (e.g. TGF-β), anaphylatoxins, glucocorticoids, catecholamines and others affecting specific subsets of acute-phase genes [7,8].

IL-6 is produced by different cells, including T cells, fibroblasts, endothelial cells, Langerhans’ cells, neutrophils, monocytes and macrophages [8,12]. In the present investigation there was no increase in initial IL-6 levels in t0 and t1. After urethrovesical anastomosis (t2) there was an increase in IL-6 in both groups (P < 0.05) but no difference in IL-6 levels between the groups (Fig. 1C). IL-6 has been shown to be a useful variable for predicting complications and organ failure after surgery [9–11]. Other authors found IL-6 to be proportional to tissue injury in animal and human studies. IL-6 has also been investigated to compare laparoscopic and open operations. In a comparison of laparoscopic vs open nephrectomy there was an increased IL-6 level in the ‘open group’[22]. In a study by Kristiansson et al.[30], several variables were determined in patients undergoing laparoscopic or open cholecystectomy. They found IL-6 to be the most sensitive variable when compared with TNF-α, IL-1β, cortisol and catecholamines, with a clear advantage in the laparoscopy group. This was confirmed by other groups, of whom some also found TNF-α and CRP to be higher after open cholecystectomy than laparoscopy [31–33]. There was no difference for laparoscopic and open hysterectomy, or laparoscopic and open hernia repair [34,35]. Only major operations seem to trigger an APR and only in these operations does the laparoscopic approach provide an advantage over the open approach.

The anti-inflammatory cytokine IL-10 is produced by monocytes and macrophages, TH2 cells and B lymphocytes [8,12]. Increased serum concentrations of IL-10 were detected in patients with septicaemia, and in parasite and mycobacterial infections, in the peritoneal dialysate of patients with peritonitis, and in the cerebrospinal fluid of children with meningitis. In the present study we found that patients in both groups had only weak changes in IL-10 serum values. Kato et al.[36] found IL-10 to be elevated during and after upper abdominal surgery, but there was no comparison with laparoscopy in that study. Brune et al.[37] showed that TH1 cytokine production was suppressed after open cholecystectomy but not after laparoscopy. IL-10 levels remained stable in both patient groups. Bellon et al.[33] also showed that IL-10 was not elevated after open and laparoscopic cholecystectomy. However, the exact role of IL-10 in quantifying operative trauma remains open.

Acute-phase proteins constitute a structurally and functionally heterogeneous group which are predominantly synthesized in the liver [7,8,13]. In the APR, levels of some (negative) acute-phase proteins decrease (e.g. pre-albumin, albumin and transferrin) while levels of other (positive) acute-phase proteins increase (e.g. CRP, SAA, haptoglobin, fibrinogen and others). Acute-phase proteins can directly neutralize inflammatory agents, help to minimize the extent of local tissue damage, and participate in tissue repair and regeneration. Although the concentrations of many components of the APR commonly increase together, not all of them increase uniformly in all patients with the same illness. CRP and SAA are two of the most important acute-phase proteins [13].

In the present study, there was an increase in the acute-phase metabolites CRP, SAA and IL-6 at 6–12 h after surgery, but no relevant increase in IL-10 levels. There was no difference in the level of these metabolites between the eLRP and RRP groups. Recent studies of nephrectomy and nephroureterectomy showed a clear advantage of laparoscopic procedures over open surgery, objectively measurable with acute-phase variables [22]. The operative approach appears to be particularly important, i.e. the surgery-associated tissue injury to skin, subcutis, and in particular, muscle tissue. In comparison, the extent of intracavitary manipulation of the target organ appears to play only a minor role [22], in addition to the use of different instruments or haemostasis, e.g. conventional coagulation or ultrasonic shears in the case of laparoscopy. By contrast with other surgical procedures, which can also be done laparoscopically, RP requires only a comparatively small access through the linea alba without damaging functional tissue (e.g. muscles) and without tissue destruction.

The generally less distinct phase after ‘aggressive’ surgery in patients having RP, and the clinically recognized more favourable course that usual, as a result, are possibly an expression of an overall weaker systemic response to the procedure than for, e.g. radical nephrectomy. If the systemic response to any surgical procedure does not reach a certain critical level, no catabolic phase occurs after ‘aggression’[8]. Thus, advantages over the conventional technique normally associated with laparoscopy are not obvious with eLRP.

An assumed and postulated lower invasiveness of eLRP could not be substantiated, based on the variables assessed in the present study; the study questions whether eLRP causes less, more or equivalent trauma than RRP, but it cannot provide a definitive answer. For the past 20 years, RRP has been used and steadily devloped; eLRP is a new technique and constantly improving. This must be considered when comparing both techniques. Thus, further multicentre studies are needed to identify the advantages of each surgical technique.

CONFLICT OF INTEREST

None declared.

Ancillary