Current Medical Research & Opinion
Article ST-0375.R1/877436 All rights reserved: reproduction in whole or part not permitted
py Un t rig au fo t ht di hor r S sp ize a la d le © y, u s 20 vi e o ew p r 14 r C o an h d p ibi om In rin ted m fo rm t a . Au e si th rc aU ng or i le is al K co ed D py us is Lim fo ers tr ite rp c i b a er n d so d ut na ow io l u nl n se oa d,
Original article Low central venous pressure versus acute normovolemic hemodilution versus conventional fluid management for reducing blood loss in radical retropubic prostatectomy: a randomized controlled trial Ashraf S. Habib
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
Judd W. Moul Thomas J. Polascik Cary N. Robertson
Department of Surgery, Duke University Medical Center, Durham, NC, USA
Anthony M. Roche
Department of Anesthesiology, University of Washington, Seattle, WA, USA
William D. White
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
Stephen E. Hill
Department of Anesthesiology, UT Southwestern Medical Center, Dallas, TX, USA
Israel Nosnick
Department of Urology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA
Co
Tong J. Gan
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
No
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
0300-7995 doi:10.1185/03007995.2013.877436
Vol. 30, No. 5, 2014, 937–943
Duke Perioperative Outcome Study Group
Address for correspondence: Ashraf S. Habib MBBCh MSc MHSc FRCA, Duke University Medical Center, Box 3094, Durham, NC 27710, USA. Tel: +1 919 668 6266; Fax: +1 919 668 6265; [email protected] Keywords: Blood loss – Central venous pressure – Fluid management – Hemodilution – Prostatectomy
Abstract
Objective: To compare acute normovolemic hemodilution versus low central venous pressure strategy versus conventional fluid management in reducing intraoperative estimated blood loss, hematocrit drop and need for blood transfusion in patients undergoing radical retropubic prostatectomy under general anesthesia. Research design and methods: Patients undergoing radical retropubic prostatectomy under general anesthesia were randomized to conventional fluid management, acute normovolemic hemodilution or low central venous pressure (5 mmHg). Treatment effects on estimated blood loss and hematocrit change were tested in multivariable regression models accounting for surgeon, prostate size, and all two-way interactions. Results: Ninety-two patients completed the study. Estimated blood loss (mean SD) was significantly lower with low central venous pressure (706 362 ml) compared to acute normovolemic hemodilution (1103 635 ml) and conventional (1051 714 ml) groups (p ¼ 0.0134). There was no difference between the groups in need for blood transfusion, or hematocrit drop from preoperative values. The multivariate model predicting estimated blood loss showed a significant effect of treatment (p ¼ 0.0028) and prostate size (p ¼ 0.0323), accounting for surgeon (p ¼ 0.0013). In the model predicting hematocrit change, accounting for surgeon difference (p ¼ 0.0037), the treatment effect depended on prostate size (p ¼ 0.0007) with the slope of low central venous pressure differing from the other two groups. Hematocrit was predicted to drop more with increased prostate size in acute normovolemic hemodilution and conventional groups but not with low central venous pressure. Key limitations: Limitations include the inability to blind providers to group assignment, possible variability between providers in estimation of blood loss, and the relatively small sample size that was not powered to detect differences between the groups in need for blood transfusion. Conclusions: Maintaining low central venous pressure reduced estimated blood loss compared to conventional fluid management and acute normovolemic hemodilution in patients undergoing radical retropubic prostatectomy but there was no difference in allogeneic blood transfusion between the groups.
Accepted: 12 December 2013; published online: 24 January 2014 Citation: Curr Med Res Opin 2014; 30:937–43
! 2014 Informa UK Ltd www.cmrojournal.com
Fluid management for prostatectomy and blood loss Habib et al.
937
Current Medical Research & Opinion Volume 30, Number 5
May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Introduction Despite advances in operative technique, blood loss during radical retropubic prostatectomy (RRP) can be of sufficient magnitude to warrant blood transfusion1,2. Studies have documented that the use of autologous blood is neither a cost effective nor an efficient practice1,3–5. Although the safety of allogeneic transfusion has improved owing to advances in screening for blood-borne pathogens, the possibility of a transfusion reaction or transfusiontransmitted disease remains a concern6. Previous studies have reported a greater rate of allogeneic transfusion among patients who receive a general anesthetic for RRP compared to those who receive regional anesthesia7,8. Some authors attributed the greater bleeding during general anesthesia as possibly due to increased central venous pressure (CVP)8. Lowering the CVP to less than 5 mmHg was found to reduce blood loss during liver surgery9–11. Such an approach has not been previously studied in patients undergoing RRP under general anesthesia. Acute normovolemic hemodilution (ANH) has also been reported to reduce allogeneic blood exposure in patients undergoing major elective surgery and RRP5. However, it is labor intensive, requires additional resources and special coordination to have trained personnel and supplies for the performance of the procedure available, and may consume additional operating room time12,13. At our institution, different strategies for fluid management during RRP have been used including conventional fluid management, ANH, and a technique maintaining low CVP. However, there was no consensus as to whether one technique is superior to the others in our practice. We therefore performed this study to compare conventional fluid management versus ANH versus maintaining CVP at 5 mmHg. We hypothesized a lower estimated blood loss with the low CVP group compared with ANH and conventional fluid management groups.
Patients and methods The study took place at Duke University Medical Center, Durham, NC, USA, between November 2005 and February 2009. After Institutional Review Board approval and written informed patient consent, American Society of Anesthesiologists physical status I–III patients scheduled to undergo RRP under general anesthesia, with a preoperative Hct level above 35%, were approached to participate in this study. The following were exclusion criteria: history of any primary hematologic disease, uncontrolled hypertension with diastolic blood pressure 4100 mmHg, unstable angina, myocardial infarction within 6 months preoperatively, autoimmune hemolysis 938
Fluid management for prostatectomy and blood loss Habib et al.
or ongoing blood loss, blood transfusion within 1 month of study entry or refusal of blood transfusion. Patients were randomized using sealed opaque envelopes to one of three treatment groups: conventional fluid management, ANH or low CVP. The randomized assignments were prepared using nQuery Advisor version 6. The assignments were made in a 1:1:1 ratio and were stratified for three surgeons, who performed all the surgeries in this study. A mixed block size was used following the algorithm in the software. All patients had an internal jugular double lumen central venous catheter and radial arterial line placed after induction of general anesthesia. CVP was measured with the patients supine before being placed in the operative position. Any difference in readings was noted and adjusted for in the data collection to reflect readings in the supine position. In the conventional fluid management group, maintenance rate of fluids was calculated as 4 ml/kg/hr for first 10 kg body weight, 2 ml/kg/hr for the second 10 kg body weight and 1 ml/kg/ hr for the remainder of body weight, and was provided using lactated ringer’s solution (LR). Insensible deficit was replaced with LR with the volume calculated as maintenance rate multiplied by the number of fasting hours. Evaporative loss was replaced with LR at 5 ml/kg/hr. Surgical blood loss was replaced using similar volumes of 6% hetastarch in balanced electrolyte solution (Hextend) up to 1 liter, then with LR on a 2:1 basis. Allogeneic blood was used as needed for Hct 525%. In the ANH group, autologous harvest volume was calculated using the formula: harvest volume ¼ (weight 70 (hemoglobin target hemoglobin))/hemoglobin, where target hemoglobin was 9 mg/dL, and weight was calculated as ideal body weight þ0.33 (actual weight - ideal body weight). There was no fluid replacement given for the first harvested unit, second and third units were replaced with equal volumes of Hextend (up to a maximum of 1 L), and the 4th unit was replaced with LR on a 2:1 ratio. Subsequent fluids were given as required to maintain mean arterial pressure 480% of baseline or460 mmHg. After removal of the prostate, harvested blood was infused in reverse order of removal at a sufficient rate to raise CVP and mean arterial pressure to baseline prior to surgical closure. In the low CVP group, maintenance fluids were administered with LR similar to the conventional management group, but there was no replacement for fasting hours. Surgical blood loss was replaced using similar volumes of Hextend up to 1 liter, then with LR on a 2:1 basis. Allogeneic blood was used for Hct525%. CVP was maintained at 5 mmHg until the prostate was removed. If CVP remained 45 mmHg, nitroglycerine infusion 5–20 mg/min was used to lower the CVP as tolerated. In all groups, phenylephrine or ephedrine was used to maintain mean arterial pressure within 20% of baseline. The anesthetic technique was standardized; patients were premedicated with midazolam IV (up to 2 mg), and www.cmrojournal.com ! 2014 Informa UK Ltd
fentanyl IV (up to 100 mg). Anesthesia was induced with propofol (1–2.5 mg/kg), and a muscle relaxant of the anesthesiologist’s choice was used to facilitate tracheal intubation. Maintenance of anesthesia was with isoflurane (0.5–2.5%), 50% nitrous oxide in oxygen, and fentanyl (up 5 mg/kg/h). All patients received postoperative nausea and vomiting prophylaxis using ondansetron 4 mg given within 30 minutes prior to the end of surgery. Neuromuscular blockade was reversed with neostigmine 70 mg/kg and glycopyrrolate 10 mg/kg. Postoperative analgesia was provided as per standard of care in our institution. We collected information about patients’ demographics, duration of surgery, length of hospital stay, prostate specific antigen level, prostate size at pathology, estimated blood loss, preoperative, intraoperative and postoperative Hct, intraoperative and postoperative need for allogeneic blood transfusion, amount and type of fluids given intraoperatively (crystalloids and colloids), as well as intraoperative CVP and hemodynamic data. The following laboratory measurements were also performed at baseline, end of surgery and at 24 hrs postoperatively: serum creatinine, coagulation studies, fibrinogen, and platelet function using the platelet function analyzer. Arterial blood gases were also measured at baseline after insertion of the arterial line and at the end of surgery. We also recorded adverse events for the entire duration of hospital stay. The primary end point
was intraoperative estimated blood loss. Secondary end points included change from preoperative to postoperative Hct and need for allogeneic blood transfusion.
Statistical analysis Treatment groups were compared with rank-sum tests, ANOVA, or chi-squared tests as appropriate. Treatment effects on estimated blood loss and drop in Hct were tested in multivariable linear regression models accounting for surgeon, prostate size, and all the two-way interactions with Tukey-adjusted post-hoc pairwise comparisons. Models were run on both raw data and on rank transforms to account for non-normal distributions. Pilot data from 15 patients undergoing RRP at our institution indicated a mean (SD) intraoperative blood loss with standard care of 1011 (868) ml. With 30 patients per group, this study has an 80% power to detect an effect size of 0.112 at alpha ¼ 0.05, consistent with group estimated blood loss means of, for example, 1011, 910, and 350 ml.
Results One hundred and one patients were enrolled. Of those, 92 patients completed the study and were included in the analysis. The flow of patients in the study is shown in Figure 1.
CONSORT Flow Diagram
Enrollment
Approached (n=120)
Excluded (n= 21) ♦ Declined to participate (n=19) ♦ Other reasons (n=2 )
Follow-Up
Allocation
Randomized (n=99)
Analysis
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5 May 2014
Allocated to ANH (n=33) ♦ Received allocated intervention (n=30) ♦ Did not receive allocated intervention (n=3, withdrew)
Allocated to Conventional (n=33) ♦ Received allocated intervention (n=32) ♦ Did not receive allocated intervention (n=1, withdrew)
Allocated to Low CVP (n=33) ♦ Received allocated intervention (n=30) ♦ Did not receive allocated intervention (n=3, 2 withdrew, 1 could not place CVP line)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Analyzed (n=30)
Analyzed (n=32)
Analyzed (n=30)
Figure 1. CONSORT flow diagram. ANH = acute normovolemic hemodilution, CVP = central venous pressure.
! 2014 Informa UK Ltd www.cmrojournal.com
Fluid management for prostatectomy and blood loss Habib et al.
939
Current Medical Research & Opinion Volume 30, Number 5
May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
There were no significant differences between patients in the three treatment groups in age, weight, American Society of Anesthesiologists physical status, prostate specific antigen level, prostate size at pathology and preoperative Hct (Table 1). Intraoperative and postoperative data are summarized in Table 2. Duration of surgery and length of hospital stay were not significantly different between the groups. There were significant differences between the groups in the amount of crystalloids and colloids given, with the amount of both crystalloids and colloids administered being significantly lower in the low CVP group compared Table 1. Patient demographics.
Surgeon 1/2/3 Age, yrs Weight, kg ASA II/III PSA, ng/ml Prostate size, g Preoperative Hct, %
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
10/10/10 59 (7) 91 (17) 23/7 8 (6) 48 (19) 45 (3)
11/10/11 58 (8) 95 (16) 22/10 9 (15) 42 (16) 45 (3)
9/9/12 62 (6) 92 (13) 21/9 6 (3) 45 (15) 45 (3)
Data are mean (SD) or number. ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, ASA ¼ American Society of Anesthesiologists physical status, PSA ¼ prostate specific antigen.
to the other two groups. Median CVP was also significantly different between the groups being lower in the low CVP group compared with the ANH (p ¼ 0.02) and the conventional (p ¼ 0.002) groups. Median mean arterial pressure and heart rate were not significantly different between the groups. However, the amount of phenylephrine given was significantly different between the groups with the pairwise comparisons showing that patients in the ANH group received significantly more phenylephrine than those in the conventional group (p ¼ 0.005). Estimated blood loss (mean SD) was significantly lower with low CVP (706 362 ml) compared to ANH (1103 635 ml) and conventional (1051 714 ml) groups (p ¼ 0.0134). Pairwise comparisons showed that estimated blood loss was significantly lower in the low CVP group compared with the ANH (p ¼ 0.005) and conventional (p ¼ 0.03) groups with no difference between the ANH and conventional groups. There were also significant differences between the groups in Hct at the end of surgery (p ¼ 0.0149) and at hospital discharge (p ¼ 0.0432). Pairwise comparisons showed that Hct at end of surgery and at hospital discharge was significantly higher in the low CVP group compared with the conventional group (p ¼ 0.02 and p ¼ 0.03 respectively) with the other pairwise comparisons not being statistically significant. The lowest postoperative Hct, decrease from preoperative to lowest postoperative Hct, and transfusion
Table 2. Intraoperative and postoperative data.
Surgical time, min Crystalloids, ml Colloids, ml Median MAP, mmHg Median CVP, mmHg Median HR, beats per minute Phenylephrine, mg Ephedrine, mg EBL, ml EBL (surgeon 1), ml EBL (surgeon 2), ml EBL (surgeon 3), ml Lowest postoperative Hct, % Lowest postoperative Hct 525% Duration of hospital stay, days RBC transfusion (intraoperative) RBC transfusion (postoperative) Any RBC transfusion (intraoperative and postoperative) Harvest volume, ml Hct, end of surgery, % Hct, postoperative day 1, % Hct, discharge, % Lowest postoperative Hct, % Decrease from preoperative to lowest postoperative Hct, %
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
p Value
172 (60) 3492 (1759)y 950 (235)y 77 (8) 9 (6)* 71 (10) 829 (924)# 17 (18) 1103 (635)* 680 (410) 1440 (604) 1188 (658) 28 (4) 6 (20%) 2.3 (1) 0 2 2 1540 (310) 31.5 (3.2) 29.1 (4.2) 29.5 (3.7) 28.3 (4.2) 17 (4)
163 (48) 3197 (1357)y 835 (246)y 80 (8) 10 (4)* 68 (10) 289 (423) 9 (14) 1051 (714)* 627 (338) 1410 (900) 1148 (631) 28 (4) 6 (19%) 2.4 (1.8) 2 2 3 NA 31.7 (3.8)* 29.3 (4.4) 28.5 (4.6)* 27.8 (4.2) 17 (4)
160 (44) 1910 (1242) 720 (385) 77 (7) 5 (3) 68 (11) 447 (630) 14 (16) 706 (362) 622 (283) 858 (445) 654 (339) 30 (5) 4 (13%) 2.3 (0.9) 1 3 4 NA 33.4 (4.2) 31.3 (3.8) 30.9 (4) 29.9 (4.5) 16 (5)
0.6747 0.0001 0.0012 0.3801 0.0015 0.3779 0.0067 0.0628 0.0134 0.9205 0.1384 0.0505 0.1551 0.7683 0.9466 0.76 0.6619 0.5645 0.0149 0.0789 0.0432 0.1551 0.3071
Data are mean (SD) or number. ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, MAP ¼ mean arterial pressure, HR ¼ heart rate, EBL ¼ estimated blood loss, Hct ¼ hematocrit. *p50.05 versus low CVP. yp50.001 versus low CVP. #p50.01 versus conventional.
940
Fluid management for prostatectomy and blood loss Habib et al.
www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 5 May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Table 3. Laboratory values.
PREOPERATIVE VALUES Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl Base excess, mmol/l Lactate, mmol/l END OF SURGERY Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl Base excess, mmol/l Lactate, mmol/l POSTOPERATIVE DAY 1 Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
p Value
238 (52) 1 (0.1) 28.4 (4.3) 145.6 (36.7) 101.5 (22.8) 272.2 (70.5) 1.1 (0.1) 2.3 (1.8) 1 (0.3)
235 (53) 1 (0.1) 28.1 (3.6) 156 (39.4) 105.6 (28.1) 268.8 (62.7) 1.1 (0.2) 2 (2) 1.1 (0.3)
242 (48) 1 (0.1) 28.1 (3.9) 149.4 (44.7) 105.6 (28.1) 266.5 (47.6) 1.1 (0.2) 2.1 (2) 1.1 (0.3)
0.8643 0.6286 0.9770 0.5683 0.8851 0.9394 0.4195 0.7006 0.2415
178.6 (44.3) 1.1 (0.1) 28.7 (5.2) 128.9 (55.5) 90.1 (31.4) 192.4 (41.7) 1 (0.1) 1.7 (2.2) 1.5 (0.6)
183.6 (48) 1.2 (0.1) 28.8 (3.6) 132.6 (52.5) 91 (36.6) 205.4 (60.5) 1 (0.2) 0.4 (2.6) 1.4 (0.5)
197 (44) 1.1 (0.1) 27.1 (3.9) 113.5 (144) 86.5 (20.6) 217.5 (54.8) 1.1 (0.1) 0.4 (2.3) 1.6 (0.5)
0.2737 0.3251 0.1009 0.8338 0.9601 0.2174 0.0530 0.0685 0.2884
173.7 (36.6) 1.1 (0.1) 29.4 (2) 234.4 (74.5) 90.7 (24) 265.4 (63.7) 1 (0.2)
172.6 (50.8) 1.1 (0.1) 31.2 (3.3) 202.6 (91.2) 102.3 (48.5) 270.5 (65.1) 1.1 (0.2)
178.5 (39.5) 1.1 (0.1) 30 (4.8) 247.5 (76) 116.1 (59.7) 275.5 (66.5) 1.1 (0.2)
0.8469 0.6036 0.0809 0.4530 0.2274 0.8520 0.3731
Data are mean (SD). ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, INR ¼ international normalized ratio, aPTT ¼ activated partial thromboplastin time, Col/Epi ¼ collagen/epinephrine, Col/ADP ¼ collagen/ADP.
of red blood cells were not significantly different between the groups. There was also no difference between the groups in preoperative, end of surgery and postoperative day 1 coagulation tests, lactate and creatinine levels (Table 3). In the multivariate model predicting estimated blood loss, there was a significant effect of treatment (p ¼ 0.0028) and prostate size (p ¼ 0.0323) on estimated blood loss, accounting for surgeon (p ¼ 0.0013), without significant interactions between any of those terms. Pairwise comparisons showed lower estimated blood loss in the low CVP group compared with the ANH (p ¼ 0.0071) and conventional (p ¼ 0.0092) groups. In the multivariate model predicting change from preoperative to postoperative Hct, the effect of treatment depended on prostate size (prostate by treatment interaction p ¼ 0.0007), accounting for surgeon (p ¼ 0.0037). The slope of the low CVP group differed significantly from the other two groups. Specifically, the model predicted that Hct is expected to drop more as prostate size increases in the conventional and ANH groups, but not the low CVP group. The model suggests that this difference between treatments may become statistically significant in prostate sizes larger than around 55 g, although the data in that region is very sparse. ! 2014 Informa UK Ltd www.cmrojournal.com
Adverse events were reported in 14 patients (5 ANH group, 6 low CVP group and 3 conventional group). In the ANH group, two patients developed postoperative ileus, one had atelectasis with fever, one had an episode of desaturation and tachycardia and one became faint on getting out of bed on the first postoperative day. In the low CVP group one patient was taken back to the operating room for evacuation of a scrotal hematoma, two had low urine output, one had lightheadedness with hypotension, one had a vasovagal attack and one had an episode of tachycardia with desaturation that settled without treatment. In the conventional group one patient had a syncopal attack associated with hypotension, one was taken back to the operating room for evacuation of a pelvic hematoma and one developed tachycardia necessitating treatment with a beta blocker.
Discussion Maintaining CVP 5 mmHg during the intraoperative period was associated with a lower estimated blood loss compared to ANH or conventional fluid management. The change from preoperative to postoperative Hct was also significantly different between the treatment groups Fluid management for prostatectomy and blood loss Habib et al.
941
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5
May 2014
according to prostate size, with the change being greater with larger prostates in the ANH and conventional groups but not the low CVP group. Allogeneic blood transfusion carries several risks including viral transmission, hemolytic reactions and immunosuppression14. A number of studies have compared ANH with preoperative autologous donation in patients undergoing RRP under general anesthesia with regards to need for allogeneic transfusion and reported no difference between the two techniques5,15. Transfusion costs were however lower with ANH compared with preoperative autologous donation5,16. Therefore adoption of ANH was recommended based on its safety, convenience for patients, provision of fresh whole blood with functional platelets and active clotting factors as well as its lower cost compared to preoperative autologous donation12. However the technique is labor-intensive, and requires special preparation and staff experienced in its use. Furthermore, it has been suggested that ANH might prolong anesthesia and operating room time12,13. In our study, this technique was also associated with more need for vasopressor support. Maintaining a low CVP has been reported to reduce blood loss during hepatic resection by reducing pressure in the inferior vena cava, and consequently reducing hepatic venous and sinusoidal pressure17. However, such a technique was not previously investigated in patients undergoing RRP. In our study reduction of CVP was associated with lower blood loss compared to ANH and conventional fluid management, and higher Hct at end of surgery and hospital discharge compared to conventional fluid management. This is likely due to the reduction of pressure in the prostatic sinuses and periprostatic venous plexuses with the reduced CVP. Our multivariable model also suggested that this technique might be particularly beneficial in patients with large prostates. However, this result should be interpreted with caution since there were few data points in patients with large prostates. The impact of the individual surgeon on the results of radical prostatectomy has recently been highlighted suggesting that this factor might be more important that the surgical approach (open or robotic)16,18. Our study also found that the surgeon had a significant impact on the investigated outcomes. However, the group treatment effect was consistent across the three participating surgeons; the treatment by surgeon interaction term was non-significant in the two multivariable models that we investigated. This study has several limitations. It was not possible to blind the providers to the group assignment. This could have impacted the primary outcome of estimated blood loss. Blood loss was also estimated by the anesthesia provider and not measured as per standard practice in our institution; therefore variability resulting from various providers’ estimates is also a potential source of inaccuracy 942
Fluid management for prostatectomy and blood loss Habib et al.
in our primary subjective outcome. However, other more objective outcomes such as Hct at end of surgery and at hospital discharge agree with the estimated blood loss assessment. The study had a relatively small sample size and was not powered to detect differences in adverse events between the groups or in the more clinically relevant outcome of the need for blood transfusion. We also did not assess whether there was any difference in the cost associated with the three fluid management strategies.
Conclusion Maintaining CVP 5 mmHg was associated with lower estimated blood loss compared with ANH and conventional fluid management for RRP, but there was no difference in allogeneic blood transfusion between the groups. This technique might have advantages in patients with large prostates. Future large studies are needed to confirm the benefits and safety of this technique.
Transparency Declaration of funding This study was supported solely by departmental funds. Declaration of financial/other relationships A.S.H., J.W.M., T.J.P., C.N.R., A.M.R., W.D.W., S.E.H., I.N., and T.J.G. have disclosed that they have no significant relationships with or financial interests in any commercial companies related to this study or article. CMRO peer reviewers may have received honoraria for their review work. The peer reviewers on this manuscript have disclosed that they have no relevant financial relationships. Acknowledgments The Duke Perioperative Outcome Study Group includes Catherine Kuhn MD, Associate Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Eugene Moretti MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Adeyemi J. Olufolabi MB, Associate Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Kerri Wahl MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC; and Christopher Young MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC. Previous Presentation: This study was presented in part at the meeting of the American Society of Anesthesiologists, San Diego, California, 16–20 October 2010.
References 1. Koch MO, Smith Jr JA. Blood loss during radical retropubic prostatectomy: is preoperative autologous blood donation indicated? J Urol 1996;156:1077-9; discussion 1079-80
www.cmrojournal.com ! 2014 Informa UK Ltd
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5 May 2014
2. Shekarriz B, Upadhyay J, Wood DP. Intraoperative, perioperative, and longterm complications of radical prostatectomy. Urol Clin North Am 2001;28:639-53 3. Goldschlag B, Afzal N, Carter HB, Fleisher LA. Is preoperative donation of autologous blood rational for radical retropubic prostatectomy? J Urol 2000;164:1968-72 4. Goh M, Kleer CG, Kielczewski P, et al. Autologous blood donation prior to anatomical radical retropubic prostatectomy: is it necessary? Urology 1997;49:569-73; discussion 574 5. Monk TG, Goodnough LT, Brecher ME, et al. A prospective randomized comparison of three blood conservation strategies for radical prostatectomy. Anesthesiology 1999;91:24-33 6. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP. Transfusion medicine. First of two parts – blood transfusion. New Engl J Med 1999;340:438-47 7. Dash A, Dunn RL, Resh J, et al. Patient, surgeon, and treatment characteristics associated with homologous blood transfusion requirement during radical retropubic prostatectomy: multivariate nomogram to assist patient counseling. Urology 2004;64:117-22 8. Shir Y, Raja SN, Frank SM, Brendler CB. Intraoperative blood loss during radical retropubic prostatectomy: epidural versus general anesthesia. Urology 1995;45:993-9 9. Jones RM, Moulton CE, Hardy KJ. Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998;85:1058-60 10. Smyrniotis V, Kostopanagiotou G, Theodoraki K, et al. The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004;187:398-402
! 2014 Informa UK Ltd www.cmrojournal.com
11. Melendez JA, Arslan V, Fischer ME, et al. Perioperative outcomes of major hepatic resections under low central venous pressure anesthesia: blood loss, blood transfusion, and the risk of postoperative renal dysfunction. J Am Col Surg 1998;187:620-5 12. Goodnough LT, Monk TG, Brecher ME. Acute normovolemic hemodilution should replace the preoperative donation of autologous blood as a method of autologous-blood procurement. Transfusion 1998;38:473-6 13. Terai A, Terada N, Yoshimura K, et al. Use of acute normovolemic hemodilution in patients undergoing radical prostatectomy. Urology 2005;65:1152-6 14. Dionigi G, Boni L, Rovera F, et al. Effect of perioperative blood transfusion on clinical outcomes in hepatic surgery for cancer. World J Gastroenterol 2009;15:3976-83 15. Ness PM, Bourke DL, Walsh PC. A randomized trial of perioperative hemodilution versus transfusion of preoperatively deposited autologous blood in elective surgery. Transfusion 1992;32:226-30 16. Murphy DG, Bjartell A, Ficarra V, et al. Downsides of robot-assisted laparoscopic radical prostatectomy: limitations and complications. Eur Urol 2010;57:735-46 17. Rahbari NN, Koch M, Zimmermann JB, et al. Infrahepatic inferior vena cava clamping for reduction of central venous pressure and blood loss during hepatic resection: a randomized controlled trial. Ann Surg 2011;253:1102-10 18. Moul JW. Minimally invasive open retropubic prostatectomy: in experienced hands – still the gold standard. Oncology (Williston Park) 2012;26:610-11, 615
Fluid management for prostatectomy and blood loss Habib et al.
943
Article ST-0375.R1/877436 All rights reserved: reproduction in whole or part not permitted
py Un t rig au fo t ht di hor r S sp ize a la d le © y, u s 20 vi e o ew p r 14 r C o an h d p ibi om In rin ted m fo rm t a . Au e si th rc aU ng or i le is al K co ed D py us is Lim fo ers tr ite rp c i b a er n d so d ut na ow io l u nl n se oa d,
Original article Low central venous pressure versus acute normovolemic hemodilution versus conventional fluid management for reducing blood loss in radical retropubic prostatectomy: a randomized controlled trial Ashraf S. Habib
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
Judd W. Moul Thomas J. Polascik Cary N. Robertson
Department of Surgery, Duke University Medical Center, Durham, NC, USA
Anthony M. Roche
Department of Anesthesiology, University of Washington, Seattle, WA, USA
William D. White
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
Stephen E. Hill
Department of Anesthesiology, UT Southwestern Medical Center, Dallas, TX, USA
Israel Nosnick
Department of Urology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA
Co
Tong J. Gan
Department of Anaesthesiology, Duke University Medical Center, Durham, NC, USA
No
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
0300-7995 doi:10.1185/03007995.2013.877436
Vol. 30, No. 5, 2014, 937–943
Duke Perioperative Outcome Study Group
Address for correspondence: Ashraf S. Habib MBBCh MSc MHSc FRCA, Duke University Medical Center, Box 3094, Durham, NC 27710, USA. Tel: +1 919 668 6266; Fax: +1 919 668 6265; [email protected] Keywords: Blood loss – Central venous pressure – Fluid management – Hemodilution – Prostatectomy
Abstract
Objective: To compare acute normovolemic hemodilution versus low central venous pressure strategy versus conventional fluid management in reducing intraoperative estimated blood loss, hematocrit drop and need for blood transfusion in patients undergoing radical retropubic prostatectomy under general anesthesia. Research design and methods: Patients undergoing radical retropubic prostatectomy under general anesthesia were randomized to conventional fluid management, acute normovolemic hemodilution or low central venous pressure (5 mmHg). Treatment effects on estimated blood loss and hematocrit change were tested in multivariable regression models accounting for surgeon, prostate size, and all two-way interactions. Results: Ninety-two patients completed the study. Estimated blood loss (mean SD) was significantly lower with low central venous pressure (706 362 ml) compared to acute normovolemic hemodilution (1103 635 ml) and conventional (1051 714 ml) groups (p ¼ 0.0134). There was no difference between the groups in need for blood transfusion, or hematocrit drop from preoperative values. The multivariate model predicting estimated blood loss showed a significant effect of treatment (p ¼ 0.0028) and prostate size (p ¼ 0.0323), accounting for surgeon (p ¼ 0.0013). In the model predicting hematocrit change, accounting for surgeon difference (p ¼ 0.0037), the treatment effect depended on prostate size (p ¼ 0.0007) with the slope of low central venous pressure differing from the other two groups. Hematocrit was predicted to drop more with increased prostate size in acute normovolemic hemodilution and conventional groups but not with low central venous pressure. Key limitations: Limitations include the inability to blind providers to group assignment, possible variability between providers in estimation of blood loss, and the relatively small sample size that was not powered to detect differences between the groups in need for blood transfusion. Conclusions: Maintaining low central venous pressure reduced estimated blood loss compared to conventional fluid management and acute normovolemic hemodilution in patients undergoing radical retropubic prostatectomy but there was no difference in allogeneic blood transfusion between the groups.
Accepted: 12 December 2013; published online: 24 January 2014 Citation: Curr Med Res Opin 2014; 30:937–43
! 2014 Informa UK Ltd www.cmrojournal.com
Fluid management for prostatectomy and blood loss Habib et al.
937
Current Medical Research & Opinion Volume 30, Number 5
May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Introduction Despite advances in operative technique, blood loss during radical retropubic prostatectomy (RRP) can be of sufficient magnitude to warrant blood transfusion1,2. Studies have documented that the use of autologous blood is neither a cost effective nor an efficient practice1,3–5. Although the safety of allogeneic transfusion has improved owing to advances in screening for blood-borne pathogens, the possibility of a transfusion reaction or transfusiontransmitted disease remains a concern6. Previous studies have reported a greater rate of allogeneic transfusion among patients who receive a general anesthetic for RRP compared to those who receive regional anesthesia7,8. Some authors attributed the greater bleeding during general anesthesia as possibly due to increased central venous pressure (CVP)8. Lowering the CVP to less than 5 mmHg was found to reduce blood loss during liver surgery9–11. Such an approach has not been previously studied in patients undergoing RRP under general anesthesia. Acute normovolemic hemodilution (ANH) has also been reported to reduce allogeneic blood exposure in patients undergoing major elective surgery and RRP5. However, it is labor intensive, requires additional resources and special coordination to have trained personnel and supplies for the performance of the procedure available, and may consume additional operating room time12,13. At our institution, different strategies for fluid management during RRP have been used including conventional fluid management, ANH, and a technique maintaining low CVP. However, there was no consensus as to whether one technique is superior to the others in our practice. We therefore performed this study to compare conventional fluid management versus ANH versus maintaining CVP at 5 mmHg. We hypothesized a lower estimated blood loss with the low CVP group compared with ANH and conventional fluid management groups.
Patients and methods The study took place at Duke University Medical Center, Durham, NC, USA, between November 2005 and February 2009. After Institutional Review Board approval and written informed patient consent, American Society of Anesthesiologists physical status I–III patients scheduled to undergo RRP under general anesthesia, with a preoperative Hct level above 35%, were approached to participate in this study. The following were exclusion criteria: history of any primary hematologic disease, uncontrolled hypertension with diastolic blood pressure 4100 mmHg, unstable angina, myocardial infarction within 6 months preoperatively, autoimmune hemolysis 938
Fluid management for prostatectomy and blood loss Habib et al.
or ongoing blood loss, blood transfusion within 1 month of study entry or refusal of blood transfusion. Patients were randomized using sealed opaque envelopes to one of three treatment groups: conventional fluid management, ANH or low CVP. The randomized assignments were prepared using nQuery Advisor version 6. The assignments were made in a 1:1:1 ratio and were stratified for three surgeons, who performed all the surgeries in this study. A mixed block size was used following the algorithm in the software. All patients had an internal jugular double lumen central venous catheter and radial arterial line placed after induction of general anesthesia. CVP was measured with the patients supine before being placed in the operative position. Any difference in readings was noted and adjusted for in the data collection to reflect readings in the supine position. In the conventional fluid management group, maintenance rate of fluids was calculated as 4 ml/kg/hr for first 10 kg body weight, 2 ml/kg/hr for the second 10 kg body weight and 1 ml/kg/ hr for the remainder of body weight, and was provided using lactated ringer’s solution (LR). Insensible deficit was replaced with LR with the volume calculated as maintenance rate multiplied by the number of fasting hours. Evaporative loss was replaced with LR at 5 ml/kg/hr. Surgical blood loss was replaced using similar volumes of 6% hetastarch in balanced electrolyte solution (Hextend) up to 1 liter, then with LR on a 2:1 basis. Allogeneic blood was used as needed for Hct 525%. In the ANH group, autologous harvest volume was calculated using the formula: harvest volume ¼ (weight 70 (hemoglobin target hemoglobin))/hemoglobin, where target hemoglobin was 9 mg/dL, and weight was calculated as ideal body weight þ0.33 (actual weight - ideal body weight). There was no fluid replacement given for the first harvested unit, second and third units were replaced with equal volumes of Hextend (up to a maximum of 1 L), and the 4th unit was replaced with LR on a 2:1 ratio. Subsequent fluids were given as required to maintain mean arterial pressure 480% of baseline or460 mmHg. After removal of the prostate, harvested blood was infused in reverse order of removal at a sufficient rate to raise CVP and mean arterial pressure to baseline prior to surgical closure. In the low CVP group, maintenance fluids were administered with LR similar to the conventional management group, but there was no replacement for fasting hours. Surgical blood loss was replaced using similar volumes of Hextend up to 1 liter, then with LR on a 2:1 basis. Allogeneic blood was used for Hct525%. CVP was maintained at 5 mmHg until the prostate was removed. If CVP remained 45 mmHg, nitroglycerine infusion 5–20 mg/min was used to lower the CVP as tolerated. In all groups, phenylephrine or ephedrine was used to maintain mean arterial pressure within 20% of baseline. The anesthetic technique was standardized; patients were premedicated with midazolam IV (up to 2 mg), and www.cmrojournal.com ! 2014 Informa UK Ltd
fentanyl IV (up to 100 mg). Anesthesia was induced with propofol (1–2.5 mg/kg), and a muscle relaxant of the anesthesiologist’s choice was used to facilitate tracheal intubation. Maintenance of anesthesia was with isoflurane (0.5–2.5%), 50% nitrous oxide in oxygen, and fentanyl (up 5 mg/kg/h). All patients received postoperative nausea and vomiting prophylaxis using ondansetron 4 mg given within 30 minutes prior to the end of surgery. Neuromuscular blockade was reversed with neostigmine 70 mg/kg and glycopyrrolate 10 mg/kg. Postoperative analgesia was provided as per standard of care in our institution. We collected information about patients’ demographics, duration of surgery, length of hospital stay, prostate specific antigen level, prostate size at pathology, estimated blood loss, preoperative, intraoperative and postoperative Hct, intraoperative and postoperative need for allogeneic blood transfusion, amount and type of fluids given intraoperatively (crystalloids and colloids), as well as intraoperative CVP and hemodynamic data. The following laboratory measurements were also performed at baseline, end of surgery and at 24 hrs postoperatively: serum creatinine, coagulation studies, fibrinogen, and platelet function using the platelet function analyzer. Arterial blood gases were also measured at baseline after insertion of the arterial line and at the end of surgery. We also recorded adverse events for the entire duration of hospital stay. The primary end point
was intraoperative estimated blood loss. Secondary end points included change from preoperative to postoperative Hct and need for allogeneic blood transfusion.
Statistical analysis Treatment groups were compared with rank-sum tests, ANOVA, or chi-squared tests as appropriate. Treatment effects on estimated blood loss and drop in Hct were tested in multivariable linear regression models accounting for surgeon, prostate size, and all the two-way interactions with Tukey-adjusted post-hoc pairwise comparisons. Models were run on both raw data and on rank transforms to account for non-normal distributions. Pilot data from 15 patients undergoing RRP at our institution indicated a mean (SD) intraoperative blood loss with standard care of 1011 (868) ml. With 30 patients per group, this study has an 80% power to detect an effect size of 0.112 at alpha ¼ 0.05, consistent with group estimated blood loss means of, for example, 1011, 910, and 350 ml.
Results One hundred and one patients were enrolled. Of those, 92 patients completed the study and were included in the analysis. The flow of patients in the study is shown in Figure 1.
CONSORT Flow Diagram
Enrollment
Approached (n=120)
Excluded (n= 21) ♦ Declined to participate (n=19) ♦ Other reasons (n=2 )
Follow-Up
Allocation
Randomized (n=99)
Analysis
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5 May 2014
Allocated to ANH (n=33) ♦ Received allocated intervention (n=30) ♦ Did not receive allocated intervention (n=3, withdrew)
Allocated to Conventional (n=33) ♦ Received allocated intervention (n=32) ♦ Did not receive allocated intervention (n=1, withdrew)
Allocated to Low CVP (n=33) ♦ Received allocated intervention (n=30) ♦ Did not receive allocated intervention (n=3, 2 withdrew, 1 could not place CVP line)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Lost to follow-up (n= 0) Discontinued intervention (n= 0)
Analyzed (n=30)
Analyzed (n=32)
Analyzed (n=30)
Figure 1. CONSORT flow diagram. ANH = acute normovolemic hemodilution, CVP = central venous pressure.
! 2014 Informa UK Ltd www.cmrojournal.com
Fluid management for prostatectomy and blood loss Habib et al.
939
Current Medical Research & Opinion Volume 30, Number 5
May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
There were no significant differences between patients in the three treatment groups in age, weight, American Society of Anesthesiologists physical status, prostate specific antigen level, prostate size at pathology and preoperative Hct (Table 1). Intraoperative and postoperative data are summarized in Table 2. Duration of surgery and length of hospital stay were not significantly different between the groups. There were significant differences between the groups in the amount of crystalloids and colloids given, with the amount of both crystalloids and colloids administered being significantly lower in the low CVP group compared Table 1. Patient demographics.
Surgeon 1/2/3 Age, yrs Weight, kg ASA II/III PSA, ng/ml Prostate size, g Preoperative Hct, %
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
10/10/10 59 (7) 91 (17) 23/7 8 (6) 48 (19) 45 (3)
11/10/11 58 (8) 95 (16) 22/10 9 (15) 42 (16) 45 (3)
9/9/12 62 (6) 92 (13) 21/9 6 (3) 45 (15) 45 (3)
Data are mean (SD) or number. ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, ASA ¼ American Society of Anesthesiologists physical status, PSA ¼ prostate specific antigen.
to the other two groups. Median CVP was also significantly different between the groups being lower in the low CVP group compared with the ANH (p ¼ 0.02) and the conventional (p ¼ 0.002) groups. Median mean arterial pressure and heart rate were not significantly different between the groups. However, the amount of phenylephrine given was significantly different between the groups with the pairwise comparisons showing that patients in the ANH group received significantly more phenylephrine than those in the conventional group (p ¼ 0.005). Estimated blood loss (mean SD) was significantly lower with low CVP (706 362 ml) compared to ANH (1103 635 ml) and conventional (1051 714 ml) groups (p ¼ 0.0134). Pairwise comparisons showed that estimated blood loss was significantly lower in the low CVP group compared with the ANH (p ¼ 0.005) and conventional (p ¼ 0.03) groups with no difference between the ANH and conventional groups. There were also significant differences between the groups in Hct at the end of surgery (p ¼ 0.0149) and at hospital discharge (p ¼ 0.0432). Pairwise comparisons showed that Hct at end of surgery and at hospital discharge was significantly higher in the low CVP group compared with the conventional group (p ¼ 0.02 and p ¼ 0.03 respectively) with the other pairwise comparisons not being statistically significant. The lowest postoperative Hct, decrease from preoperative to lowest postoperative Hct, and transfusion
Table 2. Intraoperative and postoperative data.
Surgical time, min Crystalloids, ml Colloids, ml Median MAP, mmHg Median CVP, mmHg Median HR, beats per minute Phenylephrine, mg Ephedrine, mg EBL, ml EBL (surgeon 1), ml EBL (surgeon 2), ml EBL (surgeon 3), ml Lowest postoperative Hct, % Lowest postoperative Hct 525% Duration of hospital stay, days RBC transfusion (intraoperative) RBC transfusion (postoperative) Any RBC transfusion (intraoperative and postoperative) Harvest volume, ml Hct, end of surgery, % Hct, postoperative day 1, % Hct, discharge, % Lowest postoperative Hct, % Decrease from preoperative to lowest postoperative Hct, %
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
p Value
172 (60) 3492 (1759)y 950 (235)y 77 (8) 9 (6)* 71 (10) 829 (924)# 17 (18) 1103 (635)* 680 (410) 1440 (604) 1188 (658) 28 (4) 6 (20%) 2.3 (1) 0 2 2 1540 (310) 31.5 (3.2) 29.1 (4.2) 29.5 (3.7) 28.3 (4.2) 17 (4)
163 (48) 3197 (1357)y 835 (246)y 80 (8) 10 (4)* 68 (10) 289 (423) 9 (14) 1051 (714)* 627 (338) 1410 (900) 1148 (631) 28 (4) 6 (19%) 2.4 (1.8) 2 2 3 NA 31.7 (3.8)* 29.3 (4.4) 28.5 (4.6)* 27.8 (4.2) 17 (4)
160 (44) 1910 (1242) 720 (385) 77 (7) 5 (3) 68 (11) 447 (630) 14 (16) 706 (362) 622 (283) 858 (445) 654 (339) 30 (5) 4 (13%) 2.3 (0.9) 1 3 4 NA 33.4 (4.2) 31.3 (3.8) 30.9 (4) 29.9 (4.5) 16 (5)
0.6747 0.0001 0.0012 0.3801 0.0015 0.3779 0.0067 0.0628 0.0134 0.9205 0.1384 0.0505 0.1551 0.7683 0.9466 0.76 0.6619 0.5645 0.0149 0.0789 0.0432 0.1551 0.3071
Data are mean (SD) or number. ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, MAP ¼ mean arterial pressure, HR ¼ heart rate, EBL ¼ estimated blood loss, Hct ¼ hematocrit. *p50.05 versus low CVP. yp50.001 versus low CVP. #p50.01 versus conventional.
940
Fluid management for prostatectomy and blood loss Habib et al.
www.cmrojournal.com ! 2014 Informa UK Ltd
Current Medical Research & Opinion Volume 30, Number 5 May 2014
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Table 3. Laboratory values.
PREOPERATIVE VALUES Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl Base excess, mmol/l Lactate, mmol/l END OF SURGERY Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl Base excess, mmol/l Lactate, mmol/l POSTOPERATIVE DAY 1 Platelet count, 109/l INR aPTT, sec Col/Epi, sec Col/ADP, sec Fibrinogen, mg/dl Creatinine, mg/dl
ANH (n ¼ 30)
Conv (n ¼ 32)
Low CVP (n ¼ 30)
p Value
238 (52) 1 (0.1) 28.4 (4.3) 145.6 (36.7) 101.5 (22.8) 272.2 (70.5) 1.1 (0.1) 2.3 (1.8) 1 (0.3)
235 (53) 1 (0.1) 28.1 (3.6) 156 (39.4) 105.6 (28.1) 268.8 (62.7) 1.1 (0.2) 2 (2) 1.1 (0.3)
242 (48) 1 (0.1) 28.1 (3.9) 149.4 (44.7) 105.6 (28.1) 266.5 (47.6) 1.1 (0.2) 2.1 (2) 1.1 (0.3)
0.8643 0.6286 0.9770 0.5683 0.8851 0.9394 0.4195 0.7006 0.2415
178.6 (44.3) 1.1 (0.1) 28.7 (5.2) 128.9 (55.5) 90.1 (31.4) 192.4 (41.7) 1 (0.1) 1.7 (2.2) 1.5 (0.6)
183.6 (48) 1.2 (0.1) 28.8 (3.6) 132.6 (52.5) 91 (36.6) 205.4 (60.5) 1 (0.2) 0.4 (2.6) 1.4 (0.5)
197 (44) 1.1 (0.1) 27.1 (3.9) 113.5 (144) 86.5 (20.6) 217.5 (54.8) 1.1 (0.1) 0.4 (2.3) 1.6 (0.5)
0.2737 0.3251 0.1009 0.8338 0.9601 0.2174 0.0530 0.0685 0.2884
173.7 (36.6) 1.1 (0.1) 29.4 (2) 234.4 (74.5) 90.7 (24) 265.4 (63.7) 1 (0.2)
172.6 (50.8) 1.1 (0.1) 31.2 (3.3) 202.6 (91.2) 102.3 (48.5) 270.5 (65.1) 1.1 (0.2)
178.5 (39.5) 1.1 (0.1) 30 (4.8) 247.5 (76) 116.1 (59.7) 275.5 (66.5) 1.1 (0.2)
0.8469 0.6036 0.0809 0.4530 0.2274 0.8520 0.3731
Data are mean (SD). ANH ¼ acute normovolemic hemodilution, Conv ¼ conventional fluid management, CVP ¼ central venous pressure, INR ¼ international normalized ratio, aPTT ¼ activated partial thromboplastin time, Col/Epi ¼ collagen/epinephrine, Col/ADP ¼ collagen/ADP.
of red blood cells were not significantly different between the groups. There was also no difference between the groups in preoperative, end of surgery and postoperative day 1 coagulation tests, lactate and creatinine levels (Table 3). In the multivariate model predicting estimated blood loss, there was a significant effect of treatment (p ¼ 0.0028) and prostate size (p ¼ 0.0323) on estimated blood loss, accounting for surgeon (p ¼ 0.0013), without significant interactions between any of those terms. Pairwise comparisons showed lower estimated blood loss in the low CVP group compared with the ANH (p ¼ 0.0071) and conventional (p ¼ 0.0092) groups. In the multivariate model predicting change from preoperative to postoperative Hct, the effect of treatment depended on prostate size (prostate by treatment interaction p ¼ 0.0007), accounting for surgeon (p ¼ 0.0037). The slope of the low CVP group differed significantly from the other two groups. Specifically, the model predicted that Hct is expected to drop more as prostate size increases in the conventional and ANH groups, but not the low CVP group. The model suggests that this difference between treatments may become statistically significant in prostate sizes larger than around 55 g, although the data in that region is very sparse. ! 2014 Informa UK Ltd www.cmrojournal.com
Adverse events were reported in 14 patients (5 ANH group, 6 low CVP group and 3 conventional group). In the ANH group, two patients developed postoperative ileus, one had atelectasis with fever, one had an episode of desaturation and tachycardia and one became faint on getting out of bed on the first postoperative day. In the low CVP group one patient was taken back to the operating room for evacuation of a scrotal hematoma, two had low urine output, one had lightheadedness with hypotension, one had a vasovagal attack and one had an episode of tachycardia with desaturation that settled without treatment. In the conventional group one patient had a syncopal attack associated with hypotension, one was taken back to the operating room for evacuation of a pelvic hematoma and one developed tachycardia necessitating treatment with a beta blocker.
Discussion Maintaining CVP 5 mmHg during the intraoperative period was associated with a lower estimated blood loss compared to ANH or conventional fluid management. The change from preoperative to postoperative Hct was also significantly different between the treatment groups Fluid management for prostatectomy and blood loss Habib et al.
941
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5
May 2014
according to prostate size, with the change being greater with larger prostates in the ANH and conventional groups but not the low CVP group. Allogeneic blood transfusion carries several risks including viral transmission, hemolytic reactions and immunosuppression14. A number of studies have compared ANH with preoperative autologous donation in patients undergoing RRP under general anesthesia with regards to need for allogeneic transfusion and reported no difference between the two techniques5,15. Transfusion costs were however lower with ANH compared with preoperative autologous donation5,16. Therefore adoption of ANH was recommended based on its safety, convenience for patients, provision of fresh whole blood with functional platelets and active clotting factors as well as its lower cost compared to preoperative autologous donation12. However the technique is labor-intensive, and requires special preparation and staff experienced in its use. Furthermore, it has been suggested that ANH might prolong anesthesia and operating room time12,13. In our study, this technique was also associated with more need for vasopressor support. Maintaining a low CVP has been reported to reduce blood loss during hepatic resection by reducing pressure in the inferior vena cava, and consequently reducing hepatic venous and sinusoidal pressure17. However, such a technique was not previously investigated in patients undergoing RRP. In our study reduction of CVP was associated with lower blood loss compared to ANH and conventional fluid management, and higher Hct at end of surgery and hospital discharge compared to conventional fluid management. This is likely due to the reduction of pressure in the prostatic sinuses and periprostatic venous plexuses with the reduced CVP. Our multivariable model also suggested that this technique might be particularly beneficial in patients with large prostates. However, this result should be interpreted with caution since there were few data points in patients with large prostates. The impact of the individual surgeon on the results of radical prostatectomy has recently been highlighted suggesting that this factor might be more important that the surgical approach (open or robotic)16,18. Our study also found that the surgeon had a significant impact on the investigated outcomes. However, the group treatment effect was consistent across the three participating surgeons; the treatment by surgeon interaction term was non-significant in the two multivariable models that we investigated. This study has several limitations. It was not possible to blind the providers to the group assignment. This could have impacted the primary outcome of estimated blood loss. Blood loss was also estimated by the anesthesia provider and not measured as per standard practice in our institution; therefore variability resulting from various providers’ estimates is also a potential source of inaccuracy 942
Fluid management for prostatectomy and blood loss Habib et al.
in our primary subjective outcome. However, other more objective outcomes such as Hct at end of surgery and at hospital discharge agree with the estimated blood loss assessment. The study had a relatively small sample size and was not powered to detect differences in adverse events between the groups or in the more clinically relevant outcome of the need for blood transfusion. We also did not assess whether there was any difference in the cost associated with the three fluid management strategies.
Conclusion Maintaining CVP 5 mmHg was associated with lower estimated blood loss compared with ANH and conventional fluid management for RRP, but there was no difference in allogeneic blood transfusion between the groups. This technique might have advantages in patients with large prostates. Future large studies are needed to confirm the benefits and safety of this technique.
Transparency Declaration of funding This study was supported solely by departmental funds. Declaration of financial/other relationships A.S.H., J.W.M., T.J.P., C.N.R., A.M.R., W.D.W., S.E.H., I.N., and T.J.G. have disclosed that they have no significant relationships with or financial interests in any commercial companies related to this study or article. CMRO peer reviewers may have received honoraria for their review work. The peer reviewers on this manuscript have disclosed that they have no relevant financial relationships. Acknowledgments The Duke Perioperative Outcome Study Group includes Catherine Kuhn MD, Associate Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Eugene Moretti MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Adeyemi J. Olufolabi MB, Associate Professor of Anesthesiology, Duke University Medical Center, Durham, NC; Kerri Wahl MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC; and Christopher Young MD, Professor of Anesthesiology, Duke University Medical Center, Durham, NC. Previous Presentation: This study was presented in part at the meeting of the American Society of Anesthesiologists, San Diego, California, 16–20 October 2010.
References 1. Koch MO, Smith Jr JA. Blood loss during radical retropubic prostatectomy: is preoperative autologous blood donation indicated? J Urol 1996;156:1077-9; discussion 1079-80
www.cmrojournal.com ! 2014 Informa UK Ltd
Curr Med Res Opin Downloaded from informahealthcare.com by Mcgill University on 02/12/15 For personal use only.
Current Medical Research & Opinion Volume 30, Number 5 May 2014
2. Shekarriz B, Upadhyay J, Wood DP. Intraoperative, perioperative, and longterm complications of radical prostatectomy. Urol Clin North Am 2001;28:639-53 3. Goldschlag B, Afzal N, Carter HB, Fleisher LA. Is preoperative donation of autologous blood rational for radical retropubic prostatectomy? J Urol 2000;164:1968-72 4. Goh M, Kleer CG, Kielczewski P, et al. Autologous blood donation prior to anatomical radical retropubic prostatectomy: is it necessary? Urology 1997;49:569-73; discussion 574 5. Monk TG, Goodnough LT, Brecher ME, et al. A prospective randomized comparison of three blood conservation strategies for radical prostatectomy. Anesthesiology 1999;91:24-33 6. Goodnough LT, Brecher ME, Kanter MH, AuBuchon JP. Transfusion medicine. First of two parts – blood transfusion. New Engl J Med 1999;340:438-47 7. Dash A, Dunn RL, Resh J, et al. Patient, surgeon, and treatment characteristics associated with homologous blood transfusion requirement during radical retropubic prostatectomy: multivariate nomogram to assist patient counseling. Urology 2004;64:117-22 8. Shir Y, Raja SN, Frank SM, Brendler CB. Intraoperative blood loss during radical retropubic prostatectomy: epidural versus general anesthesia. Urology 1995;45:993-9 9. Jones RM, Moulton CE, Hardy KJ. Central venous pressure and its effect on blood loss during liver resection. Br J Surg 1998;85:1058-60 10. Smyrniotis V, Kostopanagiotou G, Theodoraki K, et al. The role of central venous pressure and type of vascular control in blood loss during major liver resections. Am J Surg 2004;187:398-402
! 2014 Informa UK Ltd www.cmrojournal.com
11. Melendez JA, Arslan V, Fischer ME, et al. Perioperative outcomes of major hepatic resections under low central venous pressure anesthesia: blood loss, blood transfusion, and the risk of postoperative renal dysfunction. J Am Col Surg 1998;187:620-5 12. Goodnough LT, Monk TG, Brecher ME. Acute normovolemic hemodilution should replace the preoperative donation of autologous blood as a method of autologous-blood procurement. Transfusion 1998;38:473-6 13. Terai A, Terada N, Yoshimura K, et al. Use of acute normovolemic hemodilution in patients undergoing radical prostatectomy. Urology 2005;65:1152-6 14. Dionigi G, Boni L, Rovera F, et al. Effect of perioperative blood transfusion on clinical outcomes in hepatic surgery for cancer. World J Gastroenterol 2009;15:3976-83 15. Ness PM, Bourke DL, Walsh PC. A randomized trial of perioperative hemodilution versus transfusion of preoperatively deposited autologous blood in elective surgery. Transfusion 1992;32:226-30 16. Murphy DG, Bjartell A, Ficarra V, et al. Downsides of robot-assisted laparoscopic radical prostatectomy: limitations and complications. Eur Urol 2010;57:735-46 17. Rahbari NN, Koch M, Zimmermann JB, et al. Infrahepatic inferior vena cava clamping for reduction of central venous pressure and blood loss during hepatic resection: a randomized controlled trial. Ann Surg 2011;253:1102-10 18. Moul JW. Minimally invasive open retropubic prostatectomy: in experienced hands – still the gold standard. Oncology (Williston Park) 2012;26:610-11, 615
Fluid management for prostatectomy and blood loss Habib et al.
943
