485

test is highly effective in distinguishing pregnancy-related dis-orders from other causes of low abdominal pain and bleeding.Ultrasonic examination may give further inforination on thesite of pregnancy in some but not all hCG-positive cases.

Department of Obstetricsand Gynæcology,

Jorvi District Hospital,02740 Espoo 74, Finland

EEVA-MARJA RUTANENHEIKKI TARJANNEJUHO HUOVINENMARTTI IKONEN

G6PD/MALARIA HYPOTHESIS: A BALANCED ORTRANSIENT POLYMORPHISM

SIR,-Twenty years have elapsed since Allison’ and

Motulsky2 suggested that the high gene frequency of glucose-6-phosphate dehydrogenase (G6PD) deficiency in populationsendemic for falciparum malaria might be maintained as anX-linked polymorphism by a balance of selective forces. DrMartin (Jan. 5, p. 51) proposes a mechanism based on recentclinical and in vitro studies3 by which G6PD deficiency,oxidant stress, and malaria interact. Individuals with GdB orGdB/GdB are not protected against malaria and are not suscept-ible to oxidant stress. Individuals with Gd B-lGd B- and GdB-are protected from lethal malaria but are also under negativeselection from oxidant stress. The female heterozygoteGdB/GdB-, who has both the normal and the deficient alleles,is protected from negative selection of favism and from falci-parum malaria. Martin states that this will create a situation

whereby balanced polymorphism could arise.However, the polymorphism will be balanced only if in the

GdB/GdB and GdB individuals the susceptibility to malaria isgreater than the protection against oxidant stress alone, and inthe GdB-/GdB- and GdB- individuals the susceptibility to oxi-dant stress is greater than the protection against malaria. Thatis to say, the fitnesses of the homozygotes and of the hemizy-gotes must be less than that of the heterozygotes.. If either ofthese conditions is not met, then the polymorphism will betransient with the eventual loss’of one allele and the fixationof the other. This is based on a selection model assuming anX-linked locus with two alleles, no mutation, and genotype-independent zygotic sex ratio, also assuming that total fitnessis the product of fitness components.4 The magnitudes ofopposing selective pressures, which vary from one populationto another in different environments, will determine whetherthe polymorphism will be balanced or transient.

Furthermore, Eaton5 and Friedman3 have shown that themalaria parasite itself exerts an oxidant stress on affected redcells. While this may in one sense contribute to the protectionof GdB/GdB- heterozygotes from malaria, it also adds to otheroxidant stress, such as fava beans and viral hepatitis. Thisadditional stress’ may lead to increased relative fitness ofGdB/GdB and GdB individuals and at the same time decreasedrelative fitness of GdB-/GdB- and GdB- individuals, to such anextent that a transient polymorphism with the gradual fixationof the GdB allele and the eventual loss of the GdB- allele willensue.

’

Finally, the G6PD polymorphism in many parts of Africa isa three-allele (GdA, Gd^-, GdB) polymorphism with GdA andGDA- alleles having similar geographic distributions. As yet,we have no reasonable explanation for the maintenance of theGd^ allele. While Martin offers an attractive hypothesis for themechanism of protection against malaria afforded by G6PDdeficiency, the complex interaction of infectious disease, en-vironmental factors as yet not know, and other genetic vari-

1. Allison AC. Glucose-6-phosphate dehydrogenase deficiency in red blood cellsof East Africans. Nature 1960; 186: 531-32.

2. Motulsky AG. Metabolic polymorphisms and the role of infectious disease inhuman evolution. Hum Biol 1960, 32: 28-62.

3. Friedman MJ. Oxidant damage mediates variant red cell resistance to

malaria. Nature 1979; 280: 245-47.4. Nagylaki T. Selection in one- and two-locus systems. Lecture notes in bio-

mathematics: vol. XV. Berlin: Springer-Verlag, 1977.5 Eaton JW, Eckman JR, Berger E, Jacob HS. Suppression of malaria infec-

tion by oxidant-sensitive host erythrocytes. Nature 1976; 264: 758-60.

ants should be further investigated to determine how the pres-ent pattern of distribution of the G6PD polymorphism is

produced.Comprehensive Sickle Cell Center,Department of Pathology,University of Chicago,Chicago, Illinois 60637, U.S.A.

SHELLY C. BERNSTEIN

JAMES E. BOWMAN

INADVERTENT OVERDOSE OF PARENTERALTERBUTALINE

SiR,—To our knowledge, this is only the second case-reportwhere the recommended subcutaneous dose of terbutaline

(0-25 mg) was confused with the recommended starting oraldose (2-5 mg).’ IA 35-year-old woman with insulin-requiring diabetes was

admitted to hospital with a 2 day history of lower abdominaland back pain following an amniocentesis. Her 21 week preg-nancy was complicated by large uterine fibroids. Prematurelabour was suspected and an intravenous infusion of terbuta-line was started at a rate of 0.25 mg/h. After about 12 h, dur-

ing which time her heart-rate ranged from 90 to 114/min, itwas decided to switch her to subcutaneous terbutaline 0-25

mg. However, she received an s.c. dose of 2-5 mg in error1 h after the continuous i.v. infusion of terbutaline was discon-tinued. 10 min after the s.c. dose non-radiating substernalchest pressure developed with a tachycardia of 150/min. AnECG showed inferolateral ischxmia and she was admitted tothe coronary-care unit to rule.out a myocardial infarction. Inthe coronary-care unit, examination was remarkable for apulse-rate of 150/min regular, blood-pressure 120/80 mm Hg,and a II/VI systolic ejection murmur. Her ECG revealed asinus tachycardia of 150/min with ST segment depression inleads I, II, III, F and in V4 and V6 with inverted T waves inIII and F. All medications except insulin were withheld. Hercardiac course remained essentially benign, the tachycardiaresolving after 10 h and cardiac enzymes over 2 days beingessentially normal. The ECG reverted to normal over these 2days. Terbutaline was not successful in preventing the prema-ture labour as she had a spontaneous abortion.Terbutaline, a relatively specific (32 adrenergic receptor

stimulant, is effective in reversible bronchospasm at s.c. dosesof 0-25 mg or oral doses of 2.5 mg.2 It has also shown efficacyas an investigational agent in managing premature labour,both orally and parenterally.3.4 Adverse effects are generallymild and include muscle tremor, dizziness, nervousness, palpi-tations, tachycardia, and blood-pressure changes. With paren-teral administration and larger oral doses, there is loss of

&bgr;2-receptor selectivity with greater cardiovascular side-effects. 1,5,6

Doctors and nurses must exercise caution when prescribing,or administering this drug. The parenteral preparations of ter-butaline (’Bricanyl’, Astra; ’Brethine’, Geigy) are both single-use glass ampoules containing 1 mg/ml; thus’each -ampoulecontains four normal doese of 0.25 mg each. Simply misplac-ing the decimal point can result in significant adverse effects,as in our patient. Furthermore, since the rest of each ampouleis thrown away, changing the packaging of this drug to single-dose ampoules could save lives and money.New York Hospital,New York, N.Y. 10021, U.S.A.

ROBERT D. BRANDSTETTERVINCENT GOTZ

1. Lawyer C, Pond A. Problems with terbutaline. N Engl J Med 1977; 296:821.

2. Dulfano MJ, Glass P. The bronchodilator effects of terbutaline: route of ad-ministration and patterns of response. Ann Allergy 1976; 37: 357-66.

3. Ryden G. The effect of salbutamol and terbutaline in the management ofpremature labor. Acta Obstet Gynecol Scand 1977; 56: 293-96.

4. Wallace RL, Caldwell DL, Ansbacher R, et al. Inhibition of premature laborby terbutaline. Obstet Gynecol 1978; 51: 387-92.

5. Sackner MA, Dougherty R, Watson H, et al. Hemodynamic effects of

epinephrine and terbutaline in normal man. Chest 1975; 68: 616-24.6. Smith PR, Heurich AE, Leffler CT, et al. A comparative study of subcu-

taneously administered terbutaline and epinephrine in the treatment ofacute bronchial asthma. Chest 1977; 71: 129-34.