0022-534 7/91/1462-0362$03.00/0 THt: JOURNAL OF UROLOGY Copyright© 1991 by AMERICAN UROLOGICAL ASSOCIATION, INC.
Vol. 146, 362-365, August 1991 Printed in U.S.A.
EFFECT OF TEMPORARY OCCLUSION OF HYPOGASTRIC ARTERIES ON BLOOD LOSS DURING RADICAL RETROPUBIC PROSTATECTOMY LOUIS R. KAVOUSSI, JOSEPH A. MYERS AND WILLIAM J. CATALONA From the Division of Urologic Surgery, Washington University School of Medicine, St. Louis, Missouri
ABSTRACT
We report on a prospective, partially randomized study of 130 patients, examining the effect of temporary occlusion of the hypogastric arteries on intraoperative blood loss, perioperative blood replacement and change in preoperative to postoperative hematocrit. We observed no significant difference in any of these parameters when comparing patients who did and did not undergo intraoperative occlusion of the hypogastric arteries. These findings suggest that temporary occlusion of the hypogastric arteries during radical prostatectomy does not have a major effect on the blood loss associated with this operation. Extensive collateral circulation to the prostate and a substantial venous component of blood loss may explain these findings. Banking of 3 units of autologous blood preoperatively would have decreased the need for homologous transfusions in the majority of patients. KEY WORDS: prostate, prostatectomy, arteries, carcinoma, adenocarcinoma
Adenocarcinoma of the prostate comprised 21% of all male cancers and is the most common malignancy in men.1 It is estimated in the United States that 106,000 new cases of prostate cancer will be diagnosed in 1990.1 Between 55 and 75% of all patients will be diagnosed with clinically localized tu mors.2· 3 Radical prostatectomy has been shown to be an effec tive form of treatment for these patients.4• Historically, radical prostatectomy often has been associated with considerable blood loss. Intraoperative bleeding decreases visibility in the operative field, leading to difficulty in perform ing a precise anatomical dissection. This can result in inad vertent injury to the surrounding normal structures and also incomplete tumor resection. Hemorrhage also puts the patient at risk for the adverse sequelae of hypotension. Furthermore, despite blood product screening, the patient and operating room staff are exposed to the risk of transfusion-related pathogens. In 1979 Reiner and Walsh reported on an operative technique for the management of the deep dorsal vein and Santorini's plexus that decreased intraoperative hemorrhage and improved exposure.6 However, this technique was still associated with significant blood loss, with an average of 3.5 units of blood being transfused to each patient.7 In an effort to decrease blood loss during radical prostatectomy, Peters and Walsh reported on their experience with temporary occlusion of the hypogastric arteries.7 This retrospective, nonrandomized study of patients treated with or without temporary hypogastric artery occlusion demonstrated a 40% decrease in transfusion requirement for patients in whom the hypogastric arteries were occluded. More over, an additional 30% decrease in transfused blood was dem onstrated in patients receiving regional as opposed to general anesthesia. We report a prospective, randomized study of 130 patients to determine the effect of temporary hypogastric artery occlu sion on blood loss during radical retropubic prostatectomy. 5
operation with or without intraoperative temporary occlusion of the hypogastric arteries. In some instances, owing to admin istrative error, the randomization scheme was inadvertently violated such that consecutive patients were treated with or without hypogastric artery occlusion. Thus, the study was not strictly randomized in an absolute sense. However, we do not believe that these minor protocol violations materially affected the results. Over-all, 65 patients had temporary intraoperative occlusion of both hypogastric arteries while 65 did not. When these vessels were occluded the hypogastric arteries distal to the iliac bifurcation were isolated after the pelvic node dissec tion. A vessel loop was then passed around each hypogastric artery and bulldog clamps were applied to the vessels. The clamps were placed at the takeoff of the hypogastric artery proximal to the obliterated umbilical artery. The clamps were removed after the vesicourethral anastomosis had been per formed. Preoperatively, all patients had normal coagulation studies. Three parameters to assess the effects of temporary occlusion of the hypogastric arteries on blood loss were measured: 1) intraoperative blood loss, 2) perioperative blood replacement and 3) change in preoperative to postoperative hematocrit. Intraoperative blood loss was determined by measuring blood in suction bottles and weighing laparotomy pads. Transfusions were given at the discretion of the anesthesia team based on intraoperative hemodynamic monitoring and estimated blood loss. Change in hematocrit was defined as the difference be tween the hematocrit 1 day preoperatively and 1 day postop eratively. In addition, total anesthesia time, clinical stage, pathological stage, tumor grade and intraoperative crystalloid replacement were also compared between these 2 groups. The choice of epidural or general anesthesia was at the discretion of the anesthesia team. The paired t test was used for statistical analysis.
MATERIALS AND METHODS
RESULTS
Between November 1985 and June 1988, 130 consecutive patients with clinically localized carcinoma of the prostate undergoing radical retropubic prostatectomy with pelvic lymph node dissection as described previously8 were entered into this study. All operations were performed by the same surgeon (W. J. C.). Patients were alternately randomized to undergo the
The total anesthesia time, preoperative hematocrit, postop erative hematocrit and intraoperative fluid requirement for the hypogastric artery occluded and nonoccluded groups are listed in table 1. There were no statistically significant differences between the 2 patient groups with respect to any of these parameters. The distribution of clinical stages, pathological stages and tumor grades was also comparable between the 2
Accepted for publication November 2, 1990.
362
HYPOGASTRIC ARTERY OCCLUSION AND BLOOD LOSS DURING RADICAL PROSTATECTOMY TABLE L Effects of temporary intraoperative hypogastric artery occlusion on anesthesia time and fluid losses in patients undergoing radical retropubic prostatectomy Intraop. Status of Hypogastric Arteries
Anesthesia time (hrs.)* Intraop. fluids (I.) Hematocrit (%): Preop. Postop.
Occluded (65 pts.)
Not Occluded (65 pts.)
4.7 ± 0.09 5.8 ± 4.9
4.5 ± 0.09 5.1 ± 4.9
41.9 ± 0.52 32.7 ± 0.44
41.4 ± 0.52 34.3 ± 0.47
BLOOD LOSS A
(cc) 2000,--------------------, D Occluded
500 r--------
o+------�
TRANSFUSIONS
2. Stratification of patients according to clinical and pathological parameters
Intraop. Status of Hypogastric Arteries Occluded
Not Occluded
D
6 9 28 22 0 0
2 11 24 26 1 1 2 9
D
3 10 15 18 19 0
22 3
26 30 9
20 40 5
Clinical stage: Al A2 Bl B2 C Pathological stage: Al A2 Bl B2 C Grade: Well Moderate Poor
(units)
B
Ooccluded
' o' .2 3 1_51--�- _ _ _--;
0.51-------;
14 15
groups (table 2). The parameters to assess the effect of tem porary occlusion of the hypogastric artery on blood loss for each group are shown in figure l. The mean blood loss, trans fusion requirement and change in hematocrit for the hypogas tric artery occluded group were 1,420 ml. (range 200 to 2,500 ml.), 2.2 units (range Oto 6 units) and 9.2% (range Oto 18.4%), respectively. For those without occlusion of the hypogastric arteries the mean blood loss, transfusion requirement and change in hematocrit were 1,605 ml. (range 250 to 3,800 ml.), 2.25 units (range Oto 10 units) and 7.2% (range Oto 14.9%), respectively. No statistically significant differences existed be tween these 2 groups. In the hypogastric artery occluded group 61 patients had general anesthesia, while 4 had epidural anesthesia, compared to 55 and 10, respectively, in the group not occluded. Although the number of patients receiving regional anesthesia was too small to permit definite conclusions, there were no apparent differences observed when comparing blood loss parameters in these patients with those undergoing general anesthesia (fig. 2). DISCUSSION
The development of the nerve-sparing radical retropubic prostatectomy has improved the treatment of patients with localized prostate cancer. This technique not only has had a favorable impact on postoperative impotence but also has re sulted in decreased blood loss. Unfortunately, hemorrhage re mains a potential problem even with meticulous care being exercised during the dissection. We undertook a prospective, partially (due to 3 clerical errors) randomized study to determine whether temporary in-
[§il Not Occluded
0 10001---�'-' -·'-'--
Values represent standard deviation. * Calculated from time patient entered the operating room to the time he left the room. TABLE
363
CHANGE IN HEMATOCRIT c10
(%) D Occluded
@Jl Not Occluded
FIG. 1. Influence of temporary intraoperative occlusion of hyp ogas tric arteries on blood loss (A), transfusion requirement (B) and change in hematocrit (C) during radical retropubic prostatectomy.
traoperative occlusion of the hypogastric arteries significantly decreases blood loss associated with radical retropubic prosta tectomy. We found the estimated blood loss to be comparable among patients who did and did not have temporary hyp ogas tric artery occlusion. This appeared to be owing to the fact that in these patients the postoperative hematocrit was allowed to decrease on the average of 2 hematocrit percentage units lower and they were given a mean of 0.74 I. more of crystalloid replacement. These latter 2 factors, rather than the intraoper ative blood loss, appear to account for the difference in trans fusions given. When removing the bulldog clamp after performing the anas tomosis, one could potentially worry about increased hemor rhage if, indeed, clamping significantly decreased blood flow to the operative field. We did not notice any increased bleeding when the clamps were removed as evidenced by no difference in anesthesia time, change in hematocrit or increased drainage from the Jackson Pratt drains. We believe that adequate he mostasis when securing the prostatic pedicles is more crucial to prevent blood loss and if achieved subsequent removal of the bulldog clamps will not result in further bleeding. Peters and Walsh, in a nonrandomized, retrospective analy-
364
KAVOUSSI, MYERS AND CATALONA
BLOOD LOSS
As Affected By Anesthesia Type
A 2.s
(liters) - General (116 pts.)
Occluded
- Regional (14 pts.J
Not Occluded
AU Cases
TRANSFUSIONS
As Affected By Anesthesia Type 8
5
(units) � General (116 pts.)
- Regional (14 pts.l
4>-----+----------------1
Occluded
Not Occluded
All Cases
CHANGE IN HEMATOCRI T As Affected By Anesthesia Type
Fm. 2. Influence of type of anesthesia on blood loss (A), transfusion requirement (B) and change in hematocrit ( C) during radical retropubic prostatectomy.
sis, examined the influence of temporary hypogastric artery occlusion on blood transfusion in 93 men.7 A total of 13 patients without temporary occlusion received a mean of 1.6 units of blood more than 61 who had hypogastric artery occlusion. Unfortunately, the patient groups were not treated concur rently, that is patients entering into the study early did not have temporary occlusion of the vessels, while those entering later did. With great�r experience one would expect increased expertise in performing the operation as well as alterations in operative technique that would result in improved hemostasis. The extent to which these factors influenced the results re ported by Peters and Walsh remains unquantified. The hypogastric arteries are believed to be responsible for the major blood supply to the lower urinary tract. The prostate is supplied by branches of the inferior vesical, middle hemor rhoidal and internal pudendal arteries.9 In theory, occlusion of both hypogastric arteries should decrease the blood flow to the prostate sufficiently to eliminate arterial bleeding. In reality, even with both hypogastric arteries occluded, significant arte rial bleeding can occur. The explanation for this may be related to the extensive collateral circulation that exists in the pelvis
(table 3). There are several major anastomoses among the branches of the hypogastric artery, the external iliac system and the aorta. With hypogastric artery occlusion these anas tomoses may provide a significant part of the blood supply to the bladder, seminal vesicles and prostate. Thus, temporary clamping of the hypogastric arteries may not actually decrease blood flow to these organs but, rather, it may shunt the blood through alternative pathways. Furthermore, a substantial pro portion of the blood loss during this operation comes from the venous system. Theoretically, potential complications may arise from tem porary occlusion of the hypogastric arteries. Inadvertent injury to the vessels during dissection could lead to excessive blood loss. The hypogastric vein, which lies just posterior to the artery, also may be inadvertently injured. Other potential prob lems include thrombus formation, atherosclerotic plaque dis lodgement, and injury to the ureters and obturator nerves. In our series there were no recognized complications related to isolation and occlusion of the hypogastric arteries. Another method suggested to decrease intraoperative blood loss is the use of regional rather than general anesthesia. Regional anesthesia potentially induces localized hypotension by decreasing venous tone and, thus, it may potentially decrease hemorrhage. Several physicians have demonstrated a signifi cant decrease in intraoperative blood loss in patients undergo ing hip replacements or gynecological procedures while they were under epidural anesthesia compared with general anes thesia.10-12 Peters and Walsh reported an additional 0.6 unit decrease in transfusions among patients receiving a regional anesthetic, thus, their lowest transfusion requirement occurred in patients under regional anesthesia with hypogastric artery clamping.7 Of note, this further decrease was not statistically significant. Our limited data also failed to demonstrate any advantage in decreased intraoperative blood loss with regional anesthesia. The mean postoperative hematocrit in our series was 33% and the mean number of units transfused was 2.3. Transfusion requirements could have been decreased through the use of crystalloid and volume expanders to replace blood loss, espe cially when autologous blood was not available. The importance of limiting homologous transfusions is increasing as patient concerns over acquiring blood-born pathogens escalate. In our series homologous blood transfusion could be avoided in 82% of the patients if 3 units of autologous blood were obtained preoperatively. In conclusion, we found no significant decrease in operative blood loss or the transfusion requirement with temporary oc clusion of the hypogastric arteries during radical retropubic prostatectomy. Also, the use of regional anesthesia did not reveal any beneficial trend in regard to decreased intraoperative blood loss compared with general anesthesia. The failure of hypogastric artery occlusion to decrease significantly the intra operative blood loss may be due to the extensive collateral blood supply that exists in the pelvis and to the fact that a substantial proportion of blood loss occurring with this opera tion is venous in origin. Banking of 3 or more units of autoTABLE 3.
Collateral circulation of male pelvic vasculature
Branch Obturator Sciatic Gluteal Ilio-lumbar Lateral sacral Deferential Deferential Middle hemorrhoidal
Collaterals Inferior epigastric Medial femoral circumflex Profunda femoris Sacral arteries Circumflex iliac Deep femoral Last lumbar artery Circumflex iliac Middle sacral Cremasteric Testicular Inferior mesenteric
1 HYPOGAS'T1 F�IC AR1,ERY OCCLUSION AN ) ELOOD LOSS IJUR1J\TG RADICAL P:ROSTATECTOM"i biood should eliminate the need for fusions in the majority of patients.
trans-
REFERENCES 1. Silverberg, E., Boring, C. C. and Squires, T. S.: Cancer statistics, 1990. CA, 40: 9, 1990. 2. McMillen, S. M. and Wettlaufer, J. N.: The role of repeat trans urethral biopsy in stage A carcinoma of the prostate. J. Urol., ll6: 759, 1976. 3. Murphy, G. P., Natarajan, N., Pontes, J. E., Schmitz, R. L., Smart, C. R, Schmidt, J. D. and Mettlin, C.: The national survey of prostate cancer in the United States by the American College of Surgeons. J. Urol., 127: 928, 1982. 4. Walsh, P. C. andJewett, H.J.: Radical surgery for prostatic cancer. Cancer, 45: 1906, 1980. 5. Paulson, D. F., Lin, G. H., Hinshaw, W., Stephani, S. and The Uro-Oncology Research Group: Radical surgery versus radio therapy for adenocarcinoma of the prostate. J. Urol., 128: 502, 1982. 6. Reiner, W. G. and Walsh, P. C.: An anatomical approach to the surgical management of the dorsal vein and Santorini's plexus during radical retropubic surgery.J. Urol., 121: 198, 1979. 7. Peters, C. A. and Walsh, P. C.: Blood transfusion and anesthetic practices in radical retropubic prostatectomy. J. Urol., 134: 81, 1985. 8. Catalona, W. J.: Nerve-sparing radical retropubic prostatectomy. Urol. Clin. N. Amer., 12: 187, 1985. 9. Clemente, C. D.: Gray's Anatomy, 30th ed. Philadelphia: Lea & Febiger, p. 750, 1985. 10. Keith, I.: Anaesthesia and blood loss in total hip replacement. Anaesthesia, 32: 444, 1977. 11. Modig, J. and Karlstrom, G.: Intra- and post-operative blood loss and haemodynamics in total hip replacement when performed under lumbar epidural versus general anaesthesia. Eur. J. An aesth., 4: 345, 1987. 12. Donald, J. R.: The effect of anaesthesia, hypotension, and epidural analgesia on blood loss in surgery for pelvic floor repair. Brit. J. Anaesth., 41: 155, 1969. EDITORIAL COMMENT Since our original report (reference 7 in article) we have always used bulldog clamps on the hypogastric arteries in an attempt to decrease arterial perfusion of the prostate during the procedure. Our technique differs from that used by the authors. We routinely use epidural anesthesia. Also, the bulldog clamps are removed immediately after the prostate has been removed. The operative site is inspected carefully after the bladder neck closure is performed but before the urethral anastomosis is completed because with reperfusion of the area we often find arterial bleeders that were not visible when the bulldog clamps were in place. Recently, we eliminated bulldog clamps in 50 consecutive patients. Although no patient required the transfusion of bank blood,
365
which is 225 cc blood loss in these patients was 1,400 the that in 50 consecutive patients in whom bulldog greater were left in place. These findings are similar to those reported in study, although our over-all blood loss is less. Also, subjectively, the operative field appeared less clear when bulldog clamps were elimi nated. For these reasons and because it is so easy to place bulldog damps after completion of the lymph node dissection we have contin ued to do so. Patrick C. Walsh Department of Urology The Johns Hopkins Hospital Baltimore, Maryland
REPLY BY AUTHORS We have learned that it is difficult to estimate accurately intraoper ative blood loss in patients undergoing radical prostatectomy. Once the bladder is opened, urine mixes with blood in the wound, and the combined volume is measured in the suction apparatus and on the laparotomy pads. The administration of large volumes of crystalloid and plasma expanders early in the operation causes hemodilution and diuresis, which results in a lower hematocrit and an increased volume of urine in the wound. Both of these factors can increase the estimated blood loss. Since 1985 we have studied blood loss associated with radical pros tatectomy during 3 different intervals when transfusion policies were changing. During the first interval we did not bank autologous blood and we replaced blood loss to restore the postoperative hematocrit to near normal levels. During the second interval we banked autologous blood, rendering the patients mildly anemic preoperatively, and we replaced blood to maintain the hematocrit above 30%. During this interval we usually returned all of the banked autologous blood to the patients. Most recently, we bank 4 units of autologous blood preoper atively and during the early part of the operation we infuse 3 to 5 l. crystalloid and plasma expanders. After the prostate has been removed and hemostasis has been achieved, we transfuse only enough autologous blood to maintain the hematocrit above 24%. The estimated blood loss without clamping the hypogastric arteries was 1,605 ml. during interval 1, 1,886 ml. during interval 2 and 1,575 ml. during interval 3. These differences in estimated blood loss with time in patients operated on using essentially the same technique may be related, at least partly, to changing transfusion policies. Similarly, Peters and Walsh reported a mean of 3.5 units (1,750 cc) of blood transfused without hypogastric artery clamping in 1985 (reference 7 in article) and a mean blood loss of 1,400 cc without hypogastric artery clamping approximately 5 years later in the Editorial Comment. These differences underscore the desirability of making comparisons in concurrently treated patients rather than consecutive groups of patients. We certainly agree that any simple measure that reduces the blood loss and diminishes the need for nonautologous transfusions is worth while. We have found it difficult to prove if damping the hyp ogastric arteries significantly decreases blood loss using the standard criteria for estimating operative blood loss.
Vol. 146, 362-365, August 1991 Printed in U.S.A.
EFFECT OF TEMPORARY OCCLUSION OF HYPOGASTRIC ARTERIES ON BLOOD LOSS DURING RADICAL RETROPUBIC PROSTATECTOMY LOUIS R. KAVOUSSI, JOSEPH A. MYERS AND WILLIAM J. CATALONA From the Division of Urologic Surgery, Washington University School of Medicine, St. Louis, Missouri
ABSTRACT
We report on a prospective, partially randomized study of 130 patients, examining the effect of temporary occlusion of the hypogastric arteries on intraoperative blood loss, perioperative blood replacement and change in preoperative to postoperative hematocrit. We observed no significant difference in any of these parameters when comparing patients who did and did not undergo intraoperative occlusion of the hypogastric arteries. These findings suggest that temporary occlusion of the hypogastric arteries during radical prostatectomy does not have a major effect on the blood loss associated with this operation. Extensive collateral circulation to the prostate and a substantial venous component of blood loss may explain these findings. Banking of 3 units of autologous blood preoperatively would have decreased the need for homologous transfusions in the majority of patients. KEY WORDS: prostate, prostatectomy, arteries, carcinoma, adenocarcinoma
Adenocarcinoma of the prostate comprised 21% of all male cancers and is the most common malignancy in men.1 It is estimated in the United States that 106,000 new cases of prostate cancer will be diagnosed in 1990.1 Between 55 and 75% of all patients will be diagnosed with clinically localized tu mors.2· 3 Radical prostatectomy has been shown to be an effec tive form of treatment for these patients.4• Historically, radical prostatectomy often has been associated with considerable blood loss. Intraoperative bleeding decreases visibility in the operative field, leading to difficulty in perform ing a precise anatomical dissection. This can result in inad vertent injury to the surrounding normal structures and also incomplete tumor resection. Hemorrhage also puts the patient at risk for the adverse sequelae of hypotension. Furthermore, despite blood product screening, the patient and operating room staff are exposed to the risk of transfusion-related pathogens. In 1979 Reiner and Walsh reported on an operative technique for the management of the deep dorsal vein and Santorini's plexus that decreased intraoperative hemorrhage and improved exposure.6 However, this technique was still associated with significant blood loss, with an average of 3.5 units of blood being transfused to each patient.7 In an effort to decrease blood loss during radical prostatectomy, Peters and Walsh reported on their experience with temporary occlusion of the hypogastric arteries.7 This retrospective, nonrandomized study of patients treated with or without temporary hypogastric artery occlusion demonstrated a 40% decrease in transfusion requirement for patients in whom the hypogastric arteries were occluded. More over, an additional 30% decrease in transfused blood was dem onstrated in patients receiving regional as opposed to general anesthesia. We report a prospective, randomized study of 130 patients to determine the effect of temporary hypogastric artery occlu sion on blood loss during radical retropubic prostatectomy. 5
operation with or without intraoperative temporary occlusion of the hypogastric arteries. In some instances, owing to admin istrative error, the randomization scheme was inadvertently violated such that consecutive patients were treated with or without hypogastric artery occlusion. Thus, the study was not strictly randomized in an absolute sense. However, we do not believe that these minor protocol violations materially affected the results. Over-all, 65 patients had temporary intraoperative occlusion of both hypogastric arteries while 65 did not. When these vessels were occluded the hypogastric arteries distal to the iliac bifurcation were isolated after the pelvic node dissec tion. A vessel loop was then passed around each hypogastric artery and bulldog clamps were applied to the vessels. The clamps were placed at the takeoff of the hypogastric artery proximal to the obliterated umbilical artery. The clamps were removed after the vesicourethral anastomosis had been per formed. Preoperatively, all patients had normal coagulation studies. Three parameters to assess the effects of temporary occlusion of the hypogastric arteries on blood loss were measured: 1) intraoperative blood loss, 2) perioperative blood replacement and 3) change in preoperative to postoperative hematocrit. Intraoperative blood loss was determined by measuring blood in suction bottles and weighing laparotomy pads. Transfusions were given at the discretion of the anesthesia team based on intraoperative hemodynamic monitoring and estimated blood loss. Change in hematocrit was defined as the difference be tween the hematocrit 1 day preoperatively and 1 day postop eratively. In addition, total anesthesia time, clinical stage, pathological stage, tumor grade and intraoperative crystalloid replacement were also compared between these 2 groups. The choice of epidural or general anesthesia was at the discretion of the anesthesia team. The paired t test was used for statistical analysis.
MATERIALS AND METHODS
RESULTS
Between November 1985 and June 1988, 130 consecutive patients with clinically localized carcinoma of the prostate undergoing radical retropubic prostatectomy with pelvic lymph node dissection as described previously8 were entered into this study. All operations were performed by the same surgeon (W. J. C.). Patients were alternately randomized to undergo the
The total anesthesia time, preoperative hematocrit, postop erative hematocrit and intraoperative fluid requirement for the hypogastric artery occluded and nonoccluded groups are listed in table 1. There were no statistically significant differences between the 2 patient groups with respect to any of these parameters. The distribution of clinical stages, pathological stages and tumor grades was also comparable between the 2
Accepted for publication November 2, 1990.
362
HYPOGASTRIC ARTERY OCCLUSION AND BLOOD LOSS DURING RADICAL PROSTATECTOMY TABLE L Effects of temporary intraoperative hypogastric artery occlusion on anesthesia time and fluid losses in patients undergoing radical retropubic prostatectomy Intraop. Status of Hypogastric Arteries
Anesthesia time (hrs.)* Intraop. fluids (I.) Hematocrit (%): Preop. Postop.
Occluded (65 pts.)
Not Occluded (65 pts.)
4.7 ± 0.09 5.8 ± 4.9
4.5 ± 0.09 5.1 ± 4.9
41.9 ± 0.52 32.7 ± 0.44
41.4 ± 0.52 34.3 ± 0.47
BLOOD LOSS A
(cc) 2000,--------------------, D Occluded
500 r--------
o+------�
TRANSFUSIONS
2. Stratification of patients according to clinical and pathological parameters
Intraop. Status of Hypogastric Arteries Occluded
Not Occluded
D
6 9 28 22 0 0
2 11 24 26 1 1 2 9
D
3 10 15 18 19 0
22 3
26 30 9
20 40 5
Clinical stage: Al A2 Bl B2 C Pathological stage: Al A2 Bl B2 C Grade: Well Moderate Poor
(units)
B
Ooccluded
' o' .2 3 1_51--�- _ _ _--;
0.51-------;
14 15
groups (table 2). The parameters to assess the effect of tem porary occlusion of the hypogastric artery on blood loss for each group are shown in figure l. The mean blood loss, trans fusion requirement and change in hematocrit for the hypogas tric artery occluded group were 1,420 ml. (range 200 to 2,500 ml.), 2.2 units (range Oto 6 units) and 9.2% (range Oto 18.4%), respectively. For those without occlusion of the hypogastric arteries the mean blood loss, transfusion requirement and change in hematocrit were 1,605 ml. (range 250 to 3,800 ml.), 2.25 units (range Oto 10 units) and 7.2% (range Oto 14.9%), respectively. No statistically significant differences existed be tween these 2 groups. In the hypogastric artery occluded group 61 patients had general anesthesia, while 4 had epidural anesthesia, compared to 55 and 10, respectively, in the group not occluded. Although the number of patients receiving regional anesthesia was too small to permit definite conclusions, there were no apparent differences observed when comparing blood loss parameters in these patients with those undergoing general anesthesia (fig. 2). DISCUSSION
The development of the nerve-sparing radical retropubic prostatectomy has improved the treatment of patients with localized prostate cancer. This technique not only has had a favorable impact on postoperative impotence but also has re sulted in decreased blood loss. Unfortunately, hemorrhage re mains a potential problem even with meticulous care being exercised during the dissection. We undertook a prospective, partially (due to 3 clerical errors) randomized study to determine whether temporary in-
[§il Not Occluded
0 10001---�'-' -·'-'--
Values represent standard deviation. * Calculated from time patient entered the operating room to the time he left the room. TABLE
363
CHANGE IN HEMATOCRIT c10
(%) D Occluded
@Jl Not Occluded
FIG. 1. Influence of temporary intraoperative occlusion of hyp ogas tric arteries on blood loss (A), transfusion requirement (B) and change in hematocrit (C) during radical retropubic prostatectomy.
traoperative occlusion of the hypogastric arteries significantly decreases blood loss associated with radical retropubic prosta tectomy. We found the estimated blood loss to be comparable among patients who did and did not have temporary hyp ogas tric artery occlusion. This appeared to be owing to the fact that in these patients the postoperative hematocrit was allowed to decrease on the average of 2 hematocrit percentage units lower and they were given a mean of 0.74 I. more of crystalloid replacement. These latter 2 factors, rather than the intraoper ative blood loss, appear to account for the difference in trans fusions given. When removing the bulldog clamp after performing the anas tomosis, one could potentially worry about increased hemor rhage if, indeed, clamping significantly decreased blood flow to the operative field. We did not notice any increased bleeding when the clamps were removed as evidenced by no difference in anesthesia time, change in hematocrit or increased drainage from the Jackson Pratt drains. We believe that adequate he mostasis when securing the prostatic pedicles is more crucial to prevent blood loss and if achieved subsequent removal of the bulldog clamps will not result in further bleeding. Peters and Walsh, in a nonrandomized, retrospective analy-
364
KAVOUSSI, MYERS AND CATALONA
BLOOD LOSS
As Affected By Anesthesia Type
A 2.s
(liters) - General (116 pts.)
Occluded
- Regional (14 pts.J
Not Occluded
AU Cases
TRANSFUSIONS
As Affected By Anesthesia Type 8
5
(units) � General (116 pts.)
- Regional (14 pts.l
4>-----+----------------1
Occluded
Not Occluded
All Cases
CHANGE IN HEMATOCRI T As Affected By Anesthesia Type
Fm. 2. Influence of type of anesthesia on blood loss (A), transfusion requirement (B) and change in hematocrit ( C) during radical retropubic prostatectomy.
sis, examined the influence of temporary hypogastric artery occlusion on blood transfusion in 93 men.7 A total of 13 patients without temporary occlusion received a mean of 1.6 units of blood more than 61 who had hypogastric artery occlusion. Unfortunately, the patient groups were not treated concur rently, that is patients entering into the study early did not have temporary occlusion of the vessels, while those entering later did. With great�r experience one would expect increased expertise in performing the operation as well as alterations in operative technique that would result in improved hemostasis. The extent to which these factors influenced the results re ported by Peters and Walsh remains unquantified. The hypogastric arteries are believed to be responsible for the major blood supply to the lower urinary tract. The prostate is supplied by branches of the inferior vesical, middle hemor rhoidal and internal pudendal arteries.9 In theory, occlusion of both hypogastric arteries should decrease the blood flow to the prostate sufficiently to eliminate arterial bleeding. In reality, even with both hypogastric arteries occluded, significant arte rial bleeding can occur. The explanation for this may be related to the extensive collateral circulation that exists in the pelvis
(table 3). There are several major anastomoses among the branches of the hypogastric artery, the external iliac system and the aorta. With hypogastric artery occlusion these anas tomoses may provide a significant part of the blood supply to the bladder, seminal vesicles and prostate. Thus, temporary clamping of the hypogastric arteries may not actually decrease blood flow to these organs but, rather, it may shunt the blood through alternative pathways. Furthermore, a substantial pro portion of the blood loss during this operation comes from the venous system. Theoretically, potential complications may arise from tem porary occlusion of the hypogastric arteries. Inadvertent injury to the vessels during dissection could lead to excessive blood loss. The hypogastric vein, which lies just posterior to the artery, also may be inadvertently injured. Other potential prob lems include thrombus formation, atherosclerotic plaque dis lodgement, and injury to the ureters and obturator nerves. In our series there were no recognized complications related to isolation and occlusion of the hypogastric arteries. Another method suggested to decrease intraoperative blood loss is the use of regional rather than general anesthesia. Regional anesthesia potentially induces localized hypotension by decreasing venous tone and, thus, it may potentially decrease hemorrhage. Several physicians have demonstrated a signifi cant decrease in intraoperative blood loss in patients undergo ing hip replacements or gynecological procedures while they were under epidural anesthesia compared with general anes thesia.10-12 Peters and Walsh reported an additional 0.6 unit decrease in transfusions among patients receiving a regional anesthetic, thus, their lowest transfusion requirement occurred in patients under regional anesthesia with hypogastric artery clamping.7 Of note, this further decrease was not statistically significant. Our limited data also failed to demonstrate any advantage in decreased intraoperative blood loss with regional anesthesia. The mean postoperative hematocrit in our series was 33% and the mean number of units transfused was 2.3. Transfusion requirements could have been decreased through the use of crystalloid and volume expanders to replace blood loss, espe cially when autologous blood was not available. The importance of limiting homologous transfusions is increasing as patient concerns over acquiring blood-born pathogens escalate. In our series homologous blood transfusion could be avoided in 82% of the patients if 3 units of autologous blood were obtained preoperatively. In conclusion, we found no significant decrease in operative blood loss or the transfusion requirement with temporary oc clusion of the hypogastric arteries during radical retropubic prostatectomy. Also, the use of regional anesthesia did not reveal any beneficial trend in regard to decreased intraoperative blood loss compared with general anesthesia. The failure of hypogastric artery occlusion to decrease significantly the intra operative blood loss may be due to the extensive collateral blood supply that exists in the pelvis and to the fact that a substantial proportion of blood loss occurring with this opera tion is venous in origin. Banking of 3 or more units of autoTABLE 3.
Collateral circulation of male pelvic vasculature
Branch Obturator Sciatic Gluteal Ilio-lumbar Lateral sacral Deferential Deferential Middle hemorrhoidal
Collaterals Inferior epigastric Medial femoral circumflex Profunda femoris Sacral arteries Circumflex iliac Deep femoral Last lumbar artery Circumflex iliac Middle sacral Cremasteric Testicular Inferior mesenteric
1 HYPOGAS'T1 F�IC AR1,ERY OCCLUSION AN ) ELOOD LOSS IJUR1J\TG RADICAL P:ROSTATECTOM"i biood should eliminate the need for fusions in the majority of patients.
trans-
REFERENCES 1. Silverberg, E., Boring, C. C. and Squires, T. S.: Cancer statistics, 1990. CA, 40: 9, 1990. 2. McMillen, S. M. and Wettlaufer, J. N.: The role of repeat trans urethral biopsy in stage A carcinoma of the prostate. J. Urol., ll6: 759, 1976. 3. Murphy, G. P., Natarajan, N., Pontes, J. E., Schmitz, R. L., Smart, C. R, Schmidt, J. D. and Mettlin, C.: The national survey of prostate cancer in the United States by the American College of Surgeons. J. Urol., 127: 928, 1982. 4. Walsh, P. C. andJewett, H.J.: Radical surgery for prostatic cancer. Cancer, 45: 1906, 1980. 5. Paulson, D. F., Lin, G. H., Hinshaw, W., Stephani, S. and The Uro-Oncology Research Group: Radical surgery versus radio therapy for adenocarcinoma of the prostate. J. Urol., 128: 502, 1982. 6. Reiner, W. G. and Walsh, P. C.: An anatomical approach to the surgical management of the dorsal vein and Santorini's plexus during radical retropubic surgery.J. Urol., 121: 198, 1979. 7. Peters, C. A. and Walsh, P. C.: Blood transfusion and anesthetic practices in radical retropubic prostatectomy. J. Urol., 134: 81, 1985. 8. Catalona, W. J.: Nerve-sparing radical retropubic prostatectomy. Urol. Clin. N. Amer., 12: 187, 1985. 9. Clemente, C. D.: Gray's Anatomy, 30th ed. Philadelphia: Lea & Febiger, p. 750, 1985. 10. Keith, I.: Anaesthesia and blood loss in total hip replacement. Anaesthesia, 32: 444, 1977. 11. Modig, J. and Karlstrom, G.: Intra- and post-operative blood loss and haemodynamics in total hip replacement when performed under lumbar epidural versus general anaesthesia. Eur. J. An aesth., 4: 345, 1987. 12. Donald, J. R.: The effect of anaesthesia, hypotension, and epidural analgesia on blood loss in surgery for pelvic floor repair. Brit. J. Anaesth., 41: 155, 1969. EDITORIAL COMMENT Since our original report (reference 7 in article) we have always used bulldog clamps on the hypogastric arteries in an attempt to decrease arterial perfusion of the prostate during the procedure. Our technique differs from that used by the authors. We routinely use epidural anesthesia. Also, the bulldog clamps are removed immediately after the prostate has been removed. The operative site is inspected carefully after the bladder neck closure is performed but before the urethral anastomosis is completed because with reperfusion of the area we often find arterial bleeders that were not visible when the bulldog clamps were in place. Recently, we eliminated bulldog clamps in 50 consecutive patients. Although no patient required the transfusion of bank blood,
365
which is 225 cc blood loss in these patients was 1,400 the that in 50 consecutive patients in whom bulldog greater were left in place. These findings are similar to those reported in study, although our over-all blood loss is less. Also, subjectively, the operative field appeared less clear when bulldog clamps were elimi nated. For these reasons and because it is so easy to place bulldog damps after completion of the lymph node dissection we have contin ued to do so. Patrick C. Walsh Department of Urology The Johns Hopkins Hospital Baltimore, Maryland
REPLY BY AUTHORS We have learned that it is difficult to estimate accurately intraoper ative blood loss in patients undergoing radical prostatectomy. Once the bladder is opened, urine mixes with blood in the wound, and the combined volume is measured in the suction apparatus and on the laparotomy pads. The administration of large volumes of crystalloid and plasma expanders early in the operation causes hemodilution and diuresis, which results in a lower hematocrit and an increased volume of urine in the wound. Both of these factors can increase the estimated blood loss. Since 1985 we have studied blood loss associated with radical pros tatectomy during 3 different intervals when transfusion policies were changing. During the first interval we did not bank autologous blood and we replaced blood loss to restore the postoperative hematocrit to near normal levels. During the second interval we banked autologous blood, rendering the patients mildly anemic preoperatively, and we replaced blood to maintain the hematocrit above 30%. During this interval we usually returned all of the banked autologous blood to the patients. Most recently, we bank 4 units of autologous blood preoper atively and during the early part of the operation we infuse 3 to 5 l. crystalloid and plasma expanders. After the prostate has been removed and hemostasis has been achieved, we transfuse only enough autologous blood to maintain the hematocrit above 24%. The estimated blood loss without clamping the hypogastric arteries was 1,605 ml. during interval 1, 1,886 ml. during interval 2 and 1,575 ml. during interval 3. These differences in estimated blood loss with time in patients operated on using essentially the same technique may be related, at least partly, to changing transfusion policies. Similarly, Peters and Walsh reported a mean of 3.5 units (1,750 cc) of blood transfused without hypogastric artery clamping in 1985 (reference 7 in article) and a mean blood loss of 1,400 cc without hypogastric artery clamping approximately 5 years later in the Editorial Comment. These differences underscore the desirability of making comparisons in concurrently treated patients rather than consecutive groups of patients. We certainly agree that any simple measure that reduces the blood loss and diminishes the need for nonautologous transfusions is worth while. We have found it difficult to prove if damping the hyp ogastric arteries significantly decreases blood loss using the standard criteria for estimating operative blood loss.
