- Email: [email protected]
Management of protein S deficiency
10 $0
E d i t o r i a l correspondence
frequency distribution of the data either through a scatter plot or through some measure of the variance of the data, that the authors have chosen to report interval differences after having found no difference with the average hemoglobin values. However, if the average values are equivalent for the two groups and if the <11 g m / d l hemoglobin interval shows much greater percentages in cow milk fed babies than in formula-fed babies, then there must be a higher interval (? >12 gm/dl) where cow milk-fed babies would also have a higher percentage; thus the mean values would not be appreciably different. This fact runs counterintuitive to the notion that the operating factor in the difference in the hemoglobin levels is truly the milk source. The authors state that the American Academy of Pediatrics Committee on Nutrition may need to alter its recommendations on the use of cow milk in the second 6 months of life. This m a y be true, but one cannot reach this conclusion on the basis of their article.
John B. Chessare, MD, MPH Assistant Professor o f Pediatrics Medical College o f Ohio Toledo, OH 43699
Reply To the Editor: Dr. Chessare criticizes our failure to randomly allocate infants to cow milk or commercial formula feeding, thus creating the possibility of assignment bias. Random allocation is desirable, but compliance with the elected feeding schedule was critical in this study. It was surprising to us that we had such difficulty collecting 69 infants whose parents elected to switch to whole cow milk at 6 months of age. If we had used random allocation, we probably would not have been able to complete the study. Dr. Chessare states that no information is given about hemoglobin, mean corpuscular volume, free erythrocyte protoporphyrin, and ferritin values between the two groups at the start of the study. Information regarding these measurements was deleted from the final manuscript for the sake of space. We did state that the values were not statistically different. (These values have been provided to Dr. Chessare.) We disagree with Dr. Chcssare's contention that the gross disparity in the dropout rate increases the suspicion that the groups may not have been similar from the start. The "natural" progression of feeding in non-breast-fed infants is from commercial formula to cow milk, not the reverse. Of the 24 formula-fed infants who did not complete the study, 16 switched to cow milk, four were lost to follow-up, three changed health care providers, and one desired to be dropped from the study. Two cow milk fed infants, one lost to follow-up and the other switched back to formula, were dropped from the study. Dr. Chessare suggests that we used interval differences in reporting hemoglobin values because we found no difference in average hemoglobin values between the two groups. The mean hemoglobin values at 12 months of age were not statistically different between the two groups (p = 0.057). If one looks at the
The Journal o f Pediatrics June 1988
T a b l e . H e m o g l o b i n v a l u e s in c o w m i l k - f e d a n d f o r m u l a - f e d i n f a n t s (%) a t 12 m o n t h s o f a g e <11
Cow milk Formula
gm/dl
11.0-11.5
11.6-12.0
>12.1
(%)
(%)
(%)
(%)
24.6 11.2
20.3 15.3
23.9 35.9
31.2 37.6
distribution of hemoglobin values (Table), the trend toward higher levels in formula-fed infants is clear. Finally, Dr. Chessare forgets to consider the other measures of iron insufficiency, ferritin, mean corpuscular volume, and free erythrocyte protoporphyrin in condemning this study. All values show a significant difference between the two groups at 12 months of age, indicating a higher degree of iron insufficiency in the cow milk-fed group. We believe that our study does provide enough evidence to conclude that iron supplementation should be given to infants fed cow milk in the second 6 months of life.
Walter IV. Tunnessen, Jr., MD Associate Professor o f Pediatrics and Dermatology Frank A. Oski, MD Given Professor and Chairman Department o f Pediatrics The Johns Hopkins University School o f Medicine Baltimore, MD 21205
Management of protein S deficiency To the Editor: We read with interest the article by Israels and Seshia (J PEmATR 1987;111:562-4) regarding protein C or S deficiency as a cause of childhood stroke. We question their rationale in the use of aspirin prophylaxis in these children and their concluding statement that the role of anticoagulant therapy remains to be defined in these patients. Aspirin inhibits thrombosis by acetytating platelet cyclooxygenase and thus irreversibly inhibiting the production of thromboxane A2, a potent vasoconstrictor and platelet agonist. Aspirin is effective in reducing the risk of stroke in men with transient ischemic attacks ~ and the risk of myocardial infarction in patients with unstable angina. 2 However, we are aware of no published investigations that demonstrate the efficacy of aspirin in reducing the risk of thrombosis in patients with hypercoagulable states. A recently published study 3 reviewed the clinical manifestations of 71 persons with hereditary protein S deficiency; 55% of the affected patients experienced at least one thrombotic event, and thrombosis was recurrent in 77% of these patients. The administration of warfarin sodium was successful in preventing recurrence in all patients receiving adequate therapy. This study suggests that
Volumell2 Number6
E~torialcorrespondence
appropriate management of patients with protein S deficiency who have experienced a thrombotic event (such as the patients mentioned by Israels and Seshia) should include the administration of warfarin; the role of aspirin prophylaxis as adjunctive antiplatelet therapy has not been studied in a systematic way in these patients and should be considered investigational at this time.
James A. Whitlock, MD Robert L. Janco, MD Department of Pediatrics Vanderbilt University Nashville, TN 37232 REFERENCES
1. 2.
3.
Fields WS, Lemak N A , Frankowski RF, et al. Controlled trial of aspirin in cerebral ischemia, Stroke 1977;8:301-16. Lewis HD, David JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. N Engl J Med 1983;309:396-403. Engesser L, Broekmans A W , Bri~t E, et al. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1987;106:677-82.
Reply To the Editor: The most common presentation of deficiencies of the naturally occurring anticoagulants and the hypercoagulable state is venous thrombosis, As Whitlock and Janco state, there is good evidence that warfarin anticoagulation prevents recurrent venous thromboses in this group of patients. We agree that in patients with either protein S deficiency or protein C deficiency in whom venous thrombosis develops, warfarin is presently the drug of choice. What made the patients described in our article unusual was not only their young age but the occurrence of arterial thrombotic events, which are rare but not undocumented in cases of protein S or C deficiency. One of the differences between venous thrombosis and arterial thrombosis is the major role that platelets play in arterial thrombosis? This has been the basis for the use of aspirin prophylaxis in situations of high risk of cerebral arterial thrombosis or coronary artery thrombosis. The decision to put these infants on aspirin prophylaxis was based, therefore, on the fact that these children had had arterial thrombotic episodes. Our major concern was to prevent recurrent stroke. These children have been followed up for an additional year since the article was submitted, and there have been no thrombotic events, either arterial or venous. We agree that there may be some concern as to whether these children are appropriately covered for venous thrombotic events. However, the risk that these such events wilt occur during the first two decades of life is low in these deficiencies, with a progressive risk related to age? '3 We may certainly have to reconsider our prophylactic regimen as these children grow.
S. J. lsraels, MD S. S. Seshia, MD Department of Pediatrics University o f Manitoba Winnipeg, Manitoba, Canada R3A IS1
1051
REFERENCES
1.
2. 3.
Buchanan MR. Mechanisms of pathogenesis of aterial thrombosis: potential sites of inhibition by therapeutic compounds. Semin Thromb Hemost 1988;t4:33-40_ Broekmans A W . Hereditary protein C deficiency. Haemostasis 1985; 15:233-40. Engesser L, Broekmans A W , Briet E, et al. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1987;106:677-82.
Prevalence of glucose-6-phosphate
dehydrogenase deficiency To the Editor: Mallouh and Abu-Osba (J PEDtATR 1987;111:850-2) reported that 21,6% of the male and 11% of female Saudi children had glucose-6-phosphale dehydrogenase (G6PD i deficiency when screened with the fluorescent spot test described by Beutler and Mitchell? Because G6PD deficiency is inherited as an X-linked recessive disorder, the unexpectedly high prevalence of G6PD deficiency among the female infants cannot be accounted for by the Hardy-Weinberg principle. In a study of infants with jaundice in Calgary, Alberta, Canada, 1478 infants were screened for GdPD deficiency by the same method2; 5.6% of the boys and 2.2% of the girls were G6PD deficient? Similar observations have been made by other investigators. 3'4 Several possible explanations have been suggested to account for the frequency of G6PD deficiency in female infants. Females with G6PD deficiency may be homozygous for GdPD deficiency, heterozygous but with unusually severe penetrance, or doubly heterozygous for two genes for GdPD deficiency. Alternatively, there may be a variation in the inactivation of lhe normal X chromosome. It is also possible that the screening test may be picking UP some female heterozygotes and that positive results should be confirmed by specific enzyme assays. In at least one study, 4 the diagnosis of G6PD deficiency was based on quantitative enzyme assays. Further studies are necessary before this interesting phenomenon can be elucidated. In view of the unexpected and unexplained high frequency of G6PD deficiency in female infants, we recommend that screening for GdPD deficiency be done in selected jaundiced infants, regardless of sex.
Alexander K. C. Leung, MBBS, FRCP(C) D. Ross McLeod, MD, FRCP(C) Department o f Pediatrics University o f Calgary Alberta Children's Hospital Calgary, Alberta, Canada T2T 5C7 REFERENCES
I.
2.
Beutler E, Mitchell M. Special modifications of the fluorescent screening method for glucose-6-phosphate dehydrogenase deficiency. Blood 1968;32:816-8. Leung AK. Screening of jaundiced neonates for glucose-
E d i t o r i a l correspondence
frequency distribution of the data either through a scatter plot or through some measure of the variance of the data, that the authors have chosen to report interval differences after having found no difference with the average hemoglobin values. However, if the average values are equivalent for the two groups and if the <11 g m / d l hemoglobin interval shows much greater percentages in cow milk fed babies than in formula-fed babies, then there must be a higher interval (? >12 gm/dl) where cow milk-fed babies would also have a higher percentage; thus the mean values would not be appreciably different. This fact runs counterintuitive to the notion that the operating factor in the difference in the hemoglobin levels is truly the milk source. The authors state that the American Academy of Pediatrics Committee on Nutrition may need to alter its recommendations on the use of cow milk in the second 6 months of life. This m a y be true, but one cannot reach this conclusion on the basis of their article.
John B. Chessare, MD, MPH Assistant Professor o f Pediatrics Medical College o f Ohio Toledo, OH 43699
Reply To the Editor: Dr. Chessare criticizes our failure to randomly allocate infants to cow milk or commercial formula feeding, thus creating the possibility of assignment bias. Random allocation is desirable, but compliance with the elected feeding schedule was critical in this study. It was surprising to us that we had such difficulty collecting 69 infants whose parents elected to switch to whole cow milk at 6 months of age. If we had used random allocation, we probably would not have been able to complete the study. Dr. Chessare states that no information is given about hemoglobin, mean corpuscular volume, free erythrocyte protoporphyrin, and ferritin values between the two groups at the start of the study. Information regarding these measurements was deleted from the final manuscript for the sake of space. We did state that the values were not statistically different. (These values have been provided to Dr. Chessare.) We disagree with Dr. Chcssare's contention that the gross disparity in the dropout rate increases the suspicion that the groups may not have been similar from the start. The "natural" progression of feeding in non-breast-fed infants is from commercial formula to cow milk, not the reverse. Of the 24 formula-fed infants who did not complete the study, 16 switched to cow milk, four were lost to follow-up, three changed health care providers, and one desired to be dropped from the study. Two cow milk fed infants, one lost to follow-up and the other switched back to formula, were dropped from the study. Dr. Chessare suggests that we used interval differences in reporting hemoglobin values because we found no difference in average hemoglobin values between the two groups. The mean hemoglobin values at 12 months of age were not statistically different between the two groups (p = 0.057). If one looks at the
The Journal o f Pediatrics June 1988
T a b l e . H e m o g l o b i n v a l u e s in c o w m i l k - f e d a n d f o r m u l a - f e d i n f a n t s (%) a t 12 m o n t h s o f a g e <11
Cow milk Formula
gm/dl
11.0-11.5
11.6-12.0
>12.1
(%)
(%)
(%)
(%)
24.6 11.2
20.3 15.3
23.9 35.9
31.2 37.6
distribution of hemoglobin values (Table), the trend toward higher levels in formula-fed infants is clear. Finally, Dr. Chessare forgets to consider the other measures of iron insufficiency, ferritin, mean corpuscular volume, and free erythrocyte protoporphyrin in condemning this study. All values show a significant difference between the two groups at 12 months of age, indicating a higher degree of iron insufficiency in the cow milk-fed group. We believe that our study does provide enough evidence to conclude that iron supplementation should be given to infants fed cow milk in the second 6 months of life.
Walter IV. Tunnessen, Jr., MD Associate Professor o f Pediatrics and Dermatology Frank A. Oski, MD Given Professor and Chairman Department o f Pediatrics The Johns Hopkins University School o f Medicine Baltimore, MD 21205
Management of protein S deficiency To the Editor: We read with interest the article by Israels and Seshia (J PEmATR 1987;111:562-4) regarding protein C or S deficiency as a cause of childhood stroke. We question their rationale in the use of aspirin prophylaxis in these children and their concluding statement that the role of anticoagulant therapy remains to be defined in these patients. Aspirin inhibits thrombosis by acetytating platelet cyclooxygenase and thus irreversibly inhibiting the production of thromboxane A2, a potent vasoconstrictor and platelet agonist. Aspirin is effective in reducing the risk of stroke in men with transient ischemic attacks ~ and the risk of myocardial infarction in patients with unstable angina. 2 However, we are aware of no published investigations that demonstrate the efficacy of aspirin in reducing the risk of thrombosis in patients with hypercoagulable states. A recently published study 3 reviewed the clinical manifestations of 71 persons with hereditary protein S deficiency; 55% of the affected patients experienced at least one thrombotic event, and thrombosis was recurrent in 77% of these patients. The administration of warfarin sodium was successful in preventing recurrence in all patients receiving adequate therapy. This study suggests that
Volumell2 Number6
E~torialcorrespondence
appropriate management of patients with protein S deficiency who have experienced a thrombotic event (such as the patients mentioned by Israels and Seshia) should include the administration of warfarin; the role of aspirin prophylaxis as adjunctive antiplatelet therapy has not been studied in a systematic way in these patients and should be considered investigational at this time.
James A. Whitlock, MD Robert L. Janco, MD Department of Pediatrics Vanderbilt University Nashville, TN 37232 REFERENCES
1. 2.
3.
Fields WS, Lemak N A , Frankowski RF, et al. Controlled trial of aspirin in cerebral ischemia, Stroke 1977;8:301-16. Lewis HD, David JW, Archibald DG, et al. Protective effects of aspirin against acute myocardial infarction and death in men with unstable angina. N Engl J Med 1983;309:396-403. Engesser L, Broekmans A W , Bri~t E, et al. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1987;106:677-82.
Reply To the Editor: The most common presentation of deficiencies of the naturally occurring anticoagulants and the hypercoagulable state is venous thrombosis, As Whitlock and Janco state, there is good evidence that warfarin anticoagulation prevents recurrent venous thromboses in this group of patients. We agree that in patients with either protein S deficiency or protein C deficiency in whom venous thrombosis develops, warfarin is presently the drug of choice. What made the patients described in our article unusual was not only their young age but the occurrence of arterial thrombotic events, which are rare but not undocumented in cases of protein S or C deficiency. One of the differences between venous thrombosis and arterial thrombosis is the major role that platelets play in arterial thrombosis? This has been the basis for the use of aspirin prophylaxis in situations of high risk of cerebral arterial thrombosis or coronary artery thrombosis. The decision to put these infants on aspirin prophylaxis was based, therefore, on the fact that these children had had arterial thrombotic episodes. Our major concern was to prevent recurrent stroke. These children have been followed up for an additional year since the article was submitted, and there have been no thrombotic events, either arterial or venous. We agree that there may be some concern as to whether these children are appropriately covered for venous thrombotic events. However, the risk that these such events wilt occur during the first two decades of life is low in these deficiencies, with a progressive risk related to age? '3 We may certainly have to reconsider our prophylactic regimen as these children grow.
S. J. lsraels, MD S. S. Seshia, MD Department of Pediatrics University o f Manitoba Winnipeg, Manitoba, Canada R3A IS1
1051
REFERENCES
1.
2. 3.
Buchanan MR. Mechanisms of pathogenesis of aterial thrombosis: potential sites of inhibition by therapeutic compounds. Semin Thromb Hemost 1988;t4:33-40_ Broekmans A W . Hereditary protein C deficiency. Haemostasis 1985; 15:233-40. Engesser L, Broekmans A W , Briet E, et al. Hereditary protein S deficiency: clinical manifestations. Ann Intern Med 1987;106:677-82.
Prevalence of glucose-6-phosphate
dehydrogenase deficiency To the Editor: Mallouh and Abu-Osba (J PEDtATR 1987;111:850-2) reported that 21,6% of the male and 11% of female Saudi children had glucose-6-phosphale dehydrogenase (G6PD i deficiency when screened with the fluorescent spot test described by Beutler and Mitchell? Because G6PD deficiency is inherited as an X-linked recessive disorder, the unexpectedly high prevalence of G6PD deficiency among the female infants cannot be accounted for by the Hardy-Weinberg principle. In a study of infants with jaundice in Calgary, Alberta, Canada, 1478 infants were screened for GdPD deficiency by the same method2; 5.6% of the boys and 2.2% of the girls were G6PD deficient? Similar observations have been made by other investigators. 3'4 Several possible explanations have been suggested to account for the frequency of G6PD deficiency in female infants. Females with G6PD deficiency may be homozygous for GdPD deficiency, heterozygous but with unusually severe penetrance, or doubly heterozygous for two genes for GdPD deficiency. Alternatively, there may be a variation in the inactivation of lhe normal X chromosome. It is also possible that the screening test may be picking UP some female heterozygotes and that positive results should be confirmed by specific enzyme assays. In at least one study, 4 the diagnosis of G6PD deficiency was based on quantitative enzyme assays. Further studies are necessary before this interesting phenomenon can be elucidated. In view of the unexpected and unexplained high frequency of G6PD deficiency in female infants, we recommend that screening for GdPD deficiency be done in selected jaundiced infants, regardless of sex.
Alexander K. C. Leung, MBBS, FRCP(C) D. Ross McLeod, MD, FRCP(C) Department o f Pediatrics University o f Calgary Alberta Children's Hospital Calgary, Alberta, Canada T2T 5C7 REFERENCES
I.
2.
Beutler E, Mitchell M. Special modifications of the fluorescent screening method for glucose-6-phosphate dehydrogenase deficiency. Blood 1968;32:816-8. Leung AK. Screening of jaundiced neonates for glucose-