Hematologic effects of cardiac and noncardiac surgery

Hematologic Effects of Cardiac and Noncardiac Surgery Ronald G. Pearl, MD, PhD, Robert N. Sladen, MB, MRCP (UK), FRCP (C), and Myer H. Rosenthal, MD The intraoperative and postoperative changes in number and type of WBCs in patients undergoing cardiac surgery w e r e studied. These changes were then compared with those that occurred in patients undergoing four noncardiac surgical procedures (abdominal vascular reconstruction, thoracotomy, cholecystectomy, and carotid thromboendarterectomy). Both cardiac surgery and abdominal vascular surgery resulted in a marked increase in bands and decrease

in lymphocytes. Thoracotomy and cholecystectomy resulted in similar but smaller changes. Carotid thromboendarterectomy did not produce hematologic changes. W e conclude that the hematologic changes that occur with cardiac surgery are primarily a result of the stress and trauma of major surgery rather than a result of cardiopulmonary bypass itself. © 1987 by Grune & Stratton, Inc.

ARDIAC SURGERY results in major hematologic changes including leukocytoC sis, 13 neutrophilia,4'5 increased number and per-

were obtained on the day prior to surgery. On the day of surgery, prior to induction of anesthesia, a central venous, radial arterial, and pulmonary artery catheters were inserted. All blood sampling was via the central venous catheter. CPB was performed with a Harvey bubble oxygenator at flows of approximately 2.0 L/min/m 2 and arterial pressures of 50 to 80 mm Hg. Lactated Ringer's solution was used as the bypass prime solution. Blood counts and WBC differential were obtained at the following times: prior to induction of anesthesia (PRE-IND), ten minutes on CPB (CPB-10), 60 minutes on CPB (CPB-60), end of CPB (CPB-END), chest closure (OR-END), 30 and 120 minutes in the intensive care unit (ICU-30 and ICU-120), and on postoperative days (POD) 1, 2, 3, 4, and 7.

centage of band cells? '2'4's lymphopenia,4 thrombocytopenia,6'7 and alterations in lymphocyte subpopulations and function.8'9 These changes in number and type of WBCs are associated with postoperative immunodepression and increased susceptibility to infection] '3 Noncardiac surgery also results in similar hematologic changes and postoperative immunodepression.I°q4 It is not known whether the hematologic changes that follow cardiac surgery are a result of the cardiopulmonary bypass (CPB) procedure 1'2'4 with hemodilution, mechanical trapping of leukocytes, and activation of complement, as19 or are a result of the stress and trauma of major surgery. If the changes in leukocytes are related primarily to CPB, then alterations in the CPB procedure (oxygenator type, 6'7'15'17a8pressures, flows, filters, steroids, kinin inhibitors2°) may prevent these changes and reduce postoperative immunodepression. However, if the changes in leukocytes result primarily from the stress and trauma of major surgery, then alterations in CPB are unlikely to be effective. In order to determine the relationship of changes in leukocytes to CPB, we have investigated the magnitude and timing of these changes in patients undergoing cardiac surgery and have compared these changes with those that occur in patients undergoing four other types of surgical procedures.

Noncardiac Surgery Patients Groups of ten patients undergoing one of four surgical procedures (abdominal vascular reconstruction, carotid thromboendarterectomy, cholecystectomy, thoracotomy [pneumonectomy, lobectomy, or open lung biopsy]) during the time period of the cardiac surgery study were retrospectively evaluated. Approval for the study was obtained from the Human Subjects Committee. Anesthetic technique, duration of surgery, fluid administration, and estimated blood loss were obtained from the anesthesia record. Blood count and WBC differential were recorded preoperatively and for POD 1. The abdominal vascular surgery group, in addition, had values recorded for PODs 2, 3, 4, and 7.

Anesthetic Technique The anesthetic technique was fairly uniform within groups. A high-dose narcotic-oxygen-relaxant technique supplemented with diazepam and a potent inhalation agent was used in the cardiac surgery and abdominal vascular reconstruction patients. In the other three groups, anesthesia was induced with thiopental and maintained with a potent inhalation agent and nitrous oxide, sometimes supplemented with

METHODS

Cardiac Surgery Patients Nine adult patients undergoing cardiac surgery (coronary artery bypass grafting and/or valve replacement) at Stanford University Hospital were prospectively studied following written informed consent and approval of the Human Subjects Committee. A routine blood count and differential

Prom the Department of Anesthesia, Stanford University Medical Center, CA. Address reprint requests to Ronald G. Pearl, MD, Department of Anesthesia, S-272, Stanford University Hospital, Stanford, CA 94305. © 1987 by Grune & Stratton, Inc. 0888-6296/87/0103-000453.00/0

Journal of Cardiothoracic Anesthesia, Vol I, No 3 (June), 1987: pp 205-209

205

206

PEARL, SLADEN, AND ROSENTHAL

Neutrophil counts followed the same course as total WBC counts. There was a decrease in the P M N count at the initiation of CPB (P < .05), again reflecting hemodilution; P M N counts were normal or elevated from the end of CPB through POD 7. Band counts increased markedly during CPB (P < .01) and remained elevated (approximately tenfold) from the end of CPB through POD 2; band counts then fell to normal over the next five days. There was no correlation at any time point between the band count and the duration of CPB. Lymphocyte counts decreased throughout surgery ( P < .01) and remained at approximately one fifth of normal through POD 4; lymphocyte counts returned to normal by POD 7. Monocyte counts decreased with the initiation of CPB (P < .05); monocytes returned to normal by the end of surgery and were elevated throughout the entire postoperative period (P < .05). Eosinophil counts decreased with the initiation of CPB and remained severely depressed through POD 2 (P < .05); eosinophil counts returned to normal by POD 4. Basophil counts (per 100 WBCs) were too low to be meaningful; no patient had more than 2% basophils at any time point.

low doses of a narcotic. Muscle relaxants were used in all patients. No blood conservation techniques (eg, scavenging of intraoperative blood loss, autologous blood transfusion, hemodilution) were used. A synthetic graft was always used for abdominal vascular reconstruction.

Statistics For both the cardiac surgery group and the abdominal vascular surgery group, data for each hematologic variable (eg, band count) were analyzed by a two-way repeated measures analysis of variance (subject × time point); differences between time points were determined by the protected LSD method.2~ For all five surgical groups combined, data for POD 1 were analyzed by a one-way analysis of variance. Normal hematologic values (in number × 109/L) at Stanford University Hospital at the time of the study were as follows: total WBC count, 4.0 to 11.0; neutrophil count, 2.3 to 7.0; band count, 0.0 to 0.7; lymphocyte count, 1.0 to 4.0; monocyte count, 0.1 to 0.8; eosinophil count, 0.0 to 0.5; and basophil count, 0.0 to 0.2. Surgery times were similarly analyzed by a one-way analysis of variance. Total fluid administration, blood administration, and estimated blood loss were analyzed by the Kruskal-Wallis test (a nonparametric analysis of variance). RESULTS

Cardiac Surgery Patients The hematocrit value decreased with hemodilution at the initiation of CPB and then rose throughout the course of surgery and the postoperative period. The changes that occurred in total WBC count, neutrophil (polymorphonuclear [PMN]) count, band count, lymphocyte count, monocyte count, and eosinophil count are presented in Fig 1. The total WBC count decreased with the initiation of CPB (P < .01), again reflecting hemodilution; the total WBC count was unchanged in relation to the bematocrit. The total WBC count was normal or slightly elevated from the end of CPB through POD 7. i

i

i

I

i

i

I

10,000

+

"

/l

i

~_

i

I

Noncardiac Surgery Patients Patients undergoing abdominal vascular surgery had WBC changes (Fig 1) similar to the cardiac surgery group. Band counts were elevated to approximately 2,000 from the end of surgery through POD 2 (P < .01); band counts returned to normal over the next several days. Lymphocyte counts were decreased to approximately 40% of normal from the end of surgery through POD 4 (P < .01); lymphocyte counts /t

i

ll~

O/ I I I I

I I I ,,'/ I

8,000

T

I- . . . . . . . . 0 /

I

I

I

I

I

i i i i /

J

T ~'

•. . . . . . I

I

I //

i

I

I

I

,4

i

i

i

=

I

I

// ~ '"

I

=

i

r

I

I

I

I

I

I

I

i

i

i

" ,

,

, T T-i

' -1

Fig 1.

i

0 /

//

3.ooot .......

i

.._, zO LLI

/ -i- . . . . . . . .

// /

¢

i

125 0 /

I

I

1

I

I

i~l"-4]~lf'~

I

I

II

surgery

The effects of cardiac

(0

O) a n d

abdominal

v a s c u l a r s u r g e r y (0 . . . . . - 0 ) on t o t a l W B C c o u n t , n e u t r o p h i l c o u n t (PMN), band count, lymphocyte count, monocyte count, and eosinophil count. The 13 time points (see Methods)

are not

to scale. Each d a t a p o i n t r e p r e s e n t s t h e m e a n _+1 S E M .

HEMATOLOGIC EFFECTS OF SURGERY

207

l,0,000OO0 l-

t

,000F 2°°°k ,.oo0 ~"

2,400

o°

,,ooo F

I

I

N

LU

F0 Z 0

600 [ -

000r

the cardiac surgery and abdominal vascular surgery groups. Figure 2 shows the total WBC count, P M N count, band count, lymphocyte count, and monocyte count on POD 1 for each of these five groups; eosinophil and basophil counts were too low to be meaningful. We have analyzed the data in terms of extent of operation: major (cardiac and abdominal vascular surgery), moderate (cholecystectomy and thoracotomy), and minor (carotid endarterectomy). The major extent of operation group had higher band counts (P < .01) on POD 1 than either the moderate or the minor groups; the moderate group had higher band counts (P < .05) than the minor group. Similarly, the major extent group had lower lymphocyte counts (P < .01) on POD 1 than the minor group; the moderate extent group had lymphocyte counts that were intermediate between those of the major group and those of the minor group. Total WBC counts, P M N counts, and monocyte counts were significantly higher (P < .01) in the moderate extent group than in either the major group or the minor group.

Surgical Variables O

Fig 2. Total W B C count, neutrophil (PMN) count, band count, lymphocyte count, and monocyte count on the first POD after cardiac surgery, abdominal vascular surgery, cholecystectomy, thoracotomy, or carotid thromboendarterectomy. Each bar represents the mean _+1 S E a . See the t e x t for a discussion of significant differences.

returned to normal by POD 7. There were no significant changes in monocyte counts. Eosinophil counts fell postoperatively and remained near zero through POD 1 (P < .01); eosinophil counts returned to normal by POD 4. Patients undergoing carotid thromboendarterectomy, cholecystectomy, and thoracotomy had WBC changes that differed markedly from

Duration of surgery was significantly greater for the cardiac surgery and abdominal vascular reconstruction groups than for any of the other three groups (Table 1). Similarly, the total fluid administration and blood administration were significantly greater in the cardiac surgery and abdominal vascular surgery groups than in any of the other three groups. No patient undergoing thoracotomy, cholecystectomy, or carotid thromboendarterectomy required blood. The estimated blood loss was significantly greater in the abdominal vascular reconstruction group than in the thoracotomy, cholecystectomy, and carotid thromboendarterectomy groups; blood loss was not estimated in the cardiac surgery group.

Table 1. Surgery Duration, Fluid Administration, Blood Administration, and Estimated Blood Loss for the Five Surgical Groups Cardiac Surgery duration (h) Fluid administration (L) Blood administration (units) Estimated blood loss (mL)

5.4 ± 0.5 7.4 _+ 0.9 5.0 ± 0,9 --

Abdominal 5.5 6.9 4.5 2,230

_+ 0.6 ± 0,6 _+ 0.6 _+ 253

* P < .01 compared with the cardiac surgery group. t P < .01 compared with the abdominal vascular surgery group.

Thoracotomy 2.1 1.2 0.0 288

_+ 0 . 3 " t _+ 0 . 1 " t _+ 0 . 0 " ~ -- 54 t

Cholecystectomy 1.3 0.9 0.0 181

_+ 0 . 2 " t ± 0.1" t _+ 0 . 0 " t ± 62 t

Carotid 1.8 0.7 0.0 150

+ 0.2" t _+ 0 . 1 " t _+ 0 . 0 " t _+ 22 t

208

PEARL, SLADEN, AND ROSENTHAL DISCUSSION

Despite uncomplicated intraoperative and postoperative courses, all cardiac surgery patients had marked WBC changes including an increase in bands and a decrease in lymphocytes and eosinophils. These changes began during CPB and resolved over the course of the first postoperative week. Patients undergoing abdominal vascular reconstruction had identical hematologic changes. Patients undergoing either thoracotomy or cholecystectomy showed a smaller increase in bands and a smaller decrease in lymphocytes. Our results indicate that the W B C changes that follow cardiac surgery are not unique to patients undergoing CPB. Rather, the elevation in band count and the decrease in lymphocyte count correlate primarily with the extent of operation--greatest with major extent (cardiac and abdominal vascular surgery), intermediate with moderate extent (cholecystectomy and thoracotomy), and least with minor extent (carotid thromboendarterectomy). The mediators of the 'changes that we describe are not known. Activation of complement has been demonstrated during CPB, 1519 and fragments of complement components may have major hematologic effects) 2 The effects of major noncardiac surgery on complement have not been studied. The hematologic changes that occurred were identical in the cardiac surgery and abdominal vascular surgery groups. Factors common to both groups include prolonged anesthesia and surgery, high-dose narcotic anesthetic technique, major blood loss and replacement, blood contact with foreign materials, and exten-

sive tissue trauma. Since hematologic changes were identical in both groups, the major stimulus for the hematologic changes in the cardiac surgery group should have been these factors (related to the extent of surgical operation) and not CPB itself. Futher studies are required to determine whether the hematologic effects of these factors are mediated through complement activation. The WBC and P M N counts were elevated postoperatively in the moderate extent ~f surgery group. In contrast, the major extent group showed no change in the WBC and P M N counts despite a dramatic increase in band count. The failure of the WBC and P M N counts to increase may be due to pulmonary trapping, 15 increased margination, decreased circulating half-life, or decreased production of neutrophils. These findings are similar to bacterial infections in which the WBC count is increased with moderate infection but may be normal or decreased with severe infection. The WBC changes that we describe with cardiac surgery are similar to those noted by other investigators~7; however, the onset and duration of these changes have not previously been reported. Knowledge of the magnitude and timing of WBC changes in patients undergoing surgery is important to prevent the overdiagnosis of infection in the postoperative period. 3 Our data suggest that the hematologic changes that occur in cardiac surgery patients are related primarily to the extensive surgical procedure and are unlikely to be affected by alterations solely in the CPB procedure.

REFERENCES

1. Goodman JS, Schaffner W, Collins HA, et al: Infection after cardiovascular surgery: Clinical study including examination of antimicrobial prophylaxis. N Engl J Med 278:117-123, 1968 2. WuilloudAH, Stocker FP, Weber JW: Leukocyte and differential counts after cardiac surgery in children. Helv Paediatr Acta 34:23-28, 1979 3. Bell DM, Goldmann DA, Hopkins CC, et al: Unreliability of fever and leukocytosis in the diagnosis of infection after cardiac valve surgery. J Thorac Cardiovasc Surg 75:87-90, 1978 4. Ryh/inenP, Herva E, Hollmen A, et al: Changes in peripheral blood leukocyte counts, lymphocyte subpopulations, and in vitro transformation after heart valve replacement: Effects of oxygenator type and postoperative parenteral nutrition. J Thorac CardiovascSurg 77:259-266, 1979

5. De Villota ED, Barat G, Astorqui F, et al: Pyrexia following open heart surgery:The role of bacterial infection. Anaesthesia 29:529-536, 1974 6. Kvarstein B, Cappelen C Jr, Osterud A: Blood platelets and leukocytes during cardiopulmonary bypass. Scand J Thoral CardiovascSurg 8:142-145, 1974 7. Clark RE, Beauchamp RA, Magrath RA, et al: Comparison of bubble and membrane oxygenators in short and long perfusions. J Thorac Cardiovasc Surg 78:655-666, 1979 8. Ryh~nen P, Huttunen K, Ilonen J: Natural killer cell activity after open heart surgery. Acta Anaesthesiol Stand 28:490-492, 1984 9. Eskola J, Salo M, Viljanen MK, et al: Impaired lymphocytefunctionduring openheart surgery. Br J Anaesth 56:333-338, 1984

HEMATOLOGIC EFFECTS OF SURGERY

10. Ryh~inen P: Effects of anesthesia and operative surgery on the immune response of patients of different ages. Ann Clin Res 19:7-75, 1977 11. Sato M, Eskola J: Immunosuppression after cholecystectomy. Acta Anaesthesiol Scand 21:509-516, t977 12. Culten BF, van Belle G: Lymphocyte transformation and changes in leukocyte count: Effects of anesthesia and operation. Anesthesiology 43:563-569, t975 13. Slade MS, Simmons RL, Yunis E, et al: Immunodepression after major surgery in normal patients. Surgery 78:363-372, 1975 14. Kirov SM, Shepherd J J, Donald KD: Intraoperative and postoperative changes in peripheral white blood cell counts: The contribution of stress. Aust N Z J Surg 49:738742, 1979 15. Hammerschmidt DE, Stroneek DF, Bowers TK, et al: Complement activation and neutropenia occurring during cardiopulmonary bypass. J Thorac Cardiovasc Surg 81:370-377, 1981 16. Chenoweth DE, Cooper SW, Hugli TE, et al: Complement activation during cardiopulmonary bypass: Evi-

209

dence for generation of C3a and C5a anaphylatoxins. N Engl J Med 304:497-503, 1981 17. Cavarocchi NC, Pluth JR, Schaff HV, et al: Complement activation during cardiopulmonary bypass: Comparison of bubble and membrane oxygenators. J Thorac Cardiovasc Surg 91:252-258, 1986 18. van Oeveren WV, Kazatchkine MD, DescampsLatscha B, et al: Deleterious effects of cardiopulmonary bypass: A prospective study of bubble versus membrane oxygenation. J Thorac Cardiovasc Surg 89:888-899, 1985 19. Collett B, Alhaq A, Abdullah NB, et al: Pathways to complement activation during cardiopulmonary bypass. Br Med J 289:1251-1254, 1984 20. Nagaoka H, Katori M: Inhibition of kinin formation by a kallikrein inhibitor during extracorporeal circulation in open heart surgery. Circulation 52:325-332, 1975 2i. Snedecor GW, Cochran WG: Statistical Methods. Ames, Iowa, Iowa State University Press, 1980 22. Ghebrehiwet B, Mfiller-Eberhard H J: C3e: An acidic fragment of human C3 leukocytosis-indueing activity. J Immunol 123:616-621, 1979