Correspondence

Conservative versus Aggressive Transfusion Regimens in Sickle Cell Disease

N Engl J Med 1995; 333:1641-1642December 14, 1995

Article

To the Editor:

Vichinsky and coworkers are to be applauded for their article (July 27 issue)1 reporting that patients with sickle cell disease undergoing general anesthesia and surgery fare just as well with a less intensive transfusion regimen as with a more aggressive approach. However, the authors' study did not include a “no transfusion” arm. Thus, whether some or even most patients with sickle cell disease might not require blood transfusions at all before elective surgery remains uncertain. This point is relevant not only because of the high incidence of alloimmunization and the occurrence of other transfusion-related complications in their study but also because of the substantial expense and inconvenience associated with transfusion therapy. Newer anesthetic strategies and surgical techniques (such as laparoscopic cholecystectomy), along with careful multidisciplinary patient monitoring, might render transfusions unnecessary in most patients undergoing elective surgery.

The authors somewhat misrepresented our results with a no-transfusion approach in children with sickle cell disease who require general anesthesia and surgery.2 The only complication in our small group of 10 patients who received preoperative transfusions was a hemolytic transfusion in 1 patient. Pulmonary complications of any type (including transient atelectasis) occurred after 9 of 66 surgical procedures (14 percent) in our patients who did not receive transfusions, and chest syndrome developed in four patients (6 percent) —rates similar to those observed by Vichinsky and coworkers in patients receiving transfusions. Of the 37 children in our series who had minor surgical procedures (other than tonsillectomy or adenoidectomy) without preoperative transfusions, 35 had no postoperative complications whatsoever and the remaining 2 had only transient fever.

The time has come for a randomized trial in which a less aggressive transfusion regimen similar to that described by Vichinsky and coworkers is compared with a no-transfusion approach, with all patients receiving meticulous supportive care.

George R. Buchanan, M.D.
Zora R. Rogers, M.D.
University of Texas Southwestern Medical Center at Dallas, Dallas, TX 75235

2 References

1
Vichinsky EP, Haberkern CM, Neumayr L, et al. A comparison of conservative and aggressive transfusion regimens in the perioperative management of sickle cell disease. N Engl J Med 1995;333:206-213
Full Text | Web of Science | Medline

2
Griffin TC, Buchanan GR. Elective surgery in children with sickle cell disease without preoperative blood transfusion. J Pediatr Surg 1993;28:681-685
CrossRef | Web of Science | Medline

To the Editor:

In their article on transfusions in the preoperative management of sickle cell disease, Vichinsky et al. state that comparison of the conservative- and aggressive-transfusion groups showed no significant difference in complications unrelated to transfusions. However, their Table 3 shows that the rates of postsurgical painful crises were 7 percent and 4 percent, respectively. Although this difference does not reach the 95 percent confidence level for significance with the chi-square test the authors used (t = 2.11, whereas a P value of less than 0.05 requires that t be greater than 3.84), it is significant by Student's t-test (t = 2.29, yielding P = 0.02) with sample deviations from the binomial formula [s² = frequency × (1 - frequency)]. This seemingly minor change in technique results in a major change in the central result of the paper as stated in the concluding sentence — “a conservative approach reduces the risk of transfusion-associated complications by 50 percent” — by requiring the addition of the stipulation “but increases the risk of postsurgical painful crises by 75 percent.”

As a separate matter, a possible minor error I found in the paper is the odds ratio for the category “Any complication,” given in Table 5 as 2.15; I calculate it to be 1.98.

Stephen Apfelroth, M.D., Ph.D.
New York Blood Center, New York, NY 10021

Author/Editor Response

The authors reply:

To the Editor: We agree with Drs. Buchanan and Rogers that a randomized trial comparing patients who receive transfusions with those who do not would be worthwhile. The Preoperative Transfusion Study Group has recently completed a prospective study evaluating a no-transfusion approach in 179 operations. The study analyzed risk factors such as medical history, preoperative laboratory staging, and types of surgery, including a comparison of laparoscopic to open cholecystectomy. We noted a high complication rate among patients with sickle cell anemia undergoing major surgery, regardless of preoperative transfusion regimen or type of surgery.1 Among 364 patients who underwent cholecystectomy, there was a higher incidence of serious complications in the group that did not receive transfusions than in the group that did.2 In contrast, the no-transfusion approach appears to be warranted for certain noninvasive procedures, particularly myringotomy.1 In response to the comment regarding the misrepresentation of their data: Table 3 and page 683 of their paper3 indicate that 9 of 29 patients (31 percent) undergoing thoracotomy, laparotomy, tonsillectomy, or adenoidectomy without transfusion had pulmonary complications. The other 37 patients undergoing minor procedures had no complications.

In response to Dr. Apfelroth's letter: there were a total of 11 operations with painful-crisis complications in the aggressive-transfusion group (out of 303), and 20 in the conservative-transfusion group (out of 301). The value of the chi-square test statistic (without continuity correction) is 2.8, with a significance level of 0.09. We feel comfortable in concluding that there was no significant difference, especially in the light of the large number of comparisons that were made.

With his reference to “Student's t-test . . . with sample deviations from the binomial formula,” Apfelroth is presumably referring to the test of significance of a binomial proportion, which uses a normal approximation.4 This test is algebraically equivalent to a chi-square test with the continuity correction. The resulting test statistic will be the square root of the chi-square statistic, and the P value will be identical. Therefore, Apfelroth is mistaken in his suggestion that a change of statistical technique would alter our conclusions.

The odds ratios given in Table 5 were corrected with use of the Zeger–Liang method.5 As stated in the Methods section, this correction was required because some patients had more than one operation. These values differ slightly from those that could be computed directly from the counts in the table. The odds ratio is correct as given in the table.

Elliott Vichinsky, M.D.
Lynne Neumayr, M.D.
Children's Hospital Oakland, Oakland, CA 94609

Dennis Black, Ph.D.
University of California, San Francisco, CA 94143

5 References

1
Orringer EP, Olivieri NF, Neumayr L, et al. Is preoperative transfusion always necessary in sickle cell anemia (SCA)? Blood 1995;86:Suppl:300a-300a abstract.

2
Haberkern CM, Neumayr L, Earles AN, et al. Cholecystectomy in sickle cell anemia (SCA) patients: report of transfusion study. Blood 1995;86:Suppl:142a-142a abstract.

3
Griffin TC, Buchanan GR. Elective surgery in children with sickle cell disease without preoperative blood transfusion. J Pediatr Surg 1993;28:681-685
CrossRef | Web of Science | Medline

4
Snedecor GW, Cochran WG. Statistical methods. 6th ed. Ames: Iowa State University Press, 1979:211-2.

5
Zeger SL, Liang K-Y. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121-130
CrossRef | Web of Science | Medline

Citing Articles (3)

Citing Articles

1
Ceri Hirst, Lorna Williamson, Ceri Hirst. 2012. Preoperative blood transfusions for sickle cell disease. .
CrossRef
2
Ceri Hirst, Lorna Williamson, Ceri Hirst. 2001. Preoperative blood transfusions for sickle cell disease. .
CrossRef
3
Peter Waldron, Charles Pegelow, Lynne Neumayr, Charles Haberkern, Ann Earles, Robert Wesman, Elliott Vichinsky. (1999) Tonsillectomy, Adenoidectomy, and Myringotomy in Sickle Cell Disease. Journal of Pediatric Hematology/Oncology 21:2, 129-135
CrossRef