Ectopic Pregnancies and Spontaneous in City- Incidence and Characteristics CARL L. ERHARDT, F.A.P.H.A., and HAROLD JACOBZINER, M.D., F.A.P.H.A.

pregnancies, and ectopic pregnancies Data revealed here indicate the seri- 2.6 per 1,000. Illegal abortions, re- ousness of community health's re- sponsibilities in the matter of pre- ported as such, number only five per venting fetal wastage. Though we 10,000 total reported terminated preg- need greatly to know more about but induced or causes, there is plenty to do based nancies, many illegal upon what we know. abortions were undoubtedly reported as spontaneous abortions. These figures imply that about 90 per cent of conceptions result in a live-born The history of the reporting of fetal child. It has previously been shown deaths in has been de- from New York City data 1 to be prob- scribed elsewhere. Some characteristics able that this should be no more of figure the fetal wastage have been reported than 80 per cent at the present time. and the deficiencies in the routine reg- Other investigators have reported pro- istration of fetal deaths have been as- of sessed.1-3 portions pregnancy failures ranging In the past 10 years there from 9 to 29 per cent in New York City has been a constant improvement in the for reporting specific population groups and spe- by hospitals of fetal deaths, cified types of losses.5' 8 At any rate, it and we have now reached stage the is evident from the data that a where existing discussion of specific aspects of minimum of 10 per cent of the present fetal loss seems profitable, since the data known pregnancies result in loss of the reveal differences that may suggest areas fetus, more than three-quarters of them for clinical investigation. Some facts prior to the 20th week of gestation. on therapeutic abortions in New York These losses of all City have already been early types number reported by about 15,000 a year; this figure may Tietze.4 This paper in general limits be compared with the fewer than 4,000 discussion to ectopic pregnancies and infant deaths annually. spontaneous abortions at less than 20 weeks gestation. During the three-year interval, 1952 Ectopic Pregnancies to 1954, 548,009 terminated pregnancies The figures for ectopic pregnancies of all types were reported to the New reported to the New York City Health York City Department of Health. Al- Department for the three-year period, most 90 per cent of the recorded termi- 1952 to 1954, are shown in Table 1. nations were reported as live births. A During this interval 1,413 cases were little more than 10 per cent were classi- reported, 903 white and 510 nonwhite. fied as fetal deaths. Practically all of The data are divided by age of mother, the latter were said to be spontaneous, by color, and by gravidity. Because of therapeutic abortions accounting for the small numbers involved in the higher only 2.8 per 1,000 reported terminated orders of gravidity, all women having 828 FETAL WASTAGE VOL. 46 829

Table 1-Ectopic Pregnancies by Age of Mother and by Color and Gravidity, Ratios per 1,000 Live Births, New York City, 1952-1954 White Nonwhite Gravidity Gravidity 3 3 Age of Mother Total 1 2 or More Total 1 2 or More

AllAges 2.2 2.3 1.9 2.1 6.8* 7.6 7.0 5.8 Under 20 t t t t 2.0 t t t 20-24 1.1 1.1 1.1 0.6 3.4 4.0 3.4 2.7 25-29 1.9 2.6 1.3 1.8 7.7 16.0 7.0 5.1 30-34 3.0 4.9 2.8 2.2 11.3 24.2 15.5 7.4 35-39 4.3 9.7 4.0 3.1 15.6 28.8 22.1 11.5 40andover 6.0 16.7 6.2 3.7 t t t t

* White total includes four with age and 34 with gravidity not reported; nonwhite total includes four with age and 15 with gravidity not reported. t Ratios were not computed where there were fewer than 10 caaes. the third and later pregnancies have Age of Mother-Ectopic pregnancies been combined. No rates have been occur more frequently as maternal age computed where the frequencies are increases and this is true for each gra- smaller than 10. vidity order. However, the rate of in- Gravidity in these discussions means crease by age is substantially greater the aggregate of live-born children, plus among women having first pregnancies the total fetal deaths (including abor- than it is among women having later tions) as reported on the certificates. pregnancies. For Grava I white women The reliability of the pregnancy histories the frequency of ectopic pregnancy for on these certificates may be questioned, those 40 years old or more is about 15 but the findings show consistent relation- times that among women 20- to 24- ships, regardless of possible inaccuracies years-old. But for multigravida the in the information. It is suspected, frequency for the older women is only furthermore, that some cases are still five times that among the younger group. not reported even though hospital care The pattern among nonwhites is similar, was received. Anderson, in , although the ratios are relatively high found a ratio of ectopic pregnancies to at the young ages and the proportionate each 1,000 live births of 4.5 among advance with age somewhat less than whites and 8.7 among nonwhites during among the whites. 1949-1953.7 These ratios may be com- Gravidity-The ratio among whites pared with the 2.2 and 6.8 observed in is relatively stable at about two per these New York City data. Differing age 1,000 live births for the given gravidity and gravidity patterns could account, at orders when age is disregarded. But least in part, for the differentials, but there is little reason to believe that de- Mr. Erhardt is director, Bureau of Records ficiency in reporting introduces sufficient and Statistics, and Dr. Jacobziner is assistant commissioner, Maternal and Child Health bias in the New York City data to do Services, City Department of Health, New more than lower the rates slightly for all York, N. Y. categories in this analysis. The This paper was presented before the Ameri- relation- can Association of Registration Executives ships shown by the data are unlikely to and the Statistics Section of the American be changed markedly because of any Public Health Association at the Eighty-Third Annual Meeting in Kansas City, Mo., Novem- such deficiency. ber 17, 1955. 830 JULY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH

within each age group the ratios gen- nancy histories more completely. The erally decrease with advancing numbers number of live births can be determined, of pregnancies, the decrease occurring but the type of previous pregnancy loss more sharply among older women. The cannot. From the information at hand only exception to this common pattern it cannot be determined whether ectopic is found among white women having pregnancies are repetitive. third or later pregnancies at 25-29 years Comment-It is suspected that pelvic of age. Among such women the fre- infections are a substantial, causative quency of ectopic pregnancy is higher factor in many ectopic pregnancies. In than that among Grava II women of view of the higher incidence of venereal the same age. infections among the nonwhite group, Among the nonwhites ectopic preg- therefore, the more frequent occurrence nancy occurs about three times as fre- of ectopic pregnancy among them does quently as among the whites. The gen- not appear illogical. Furthermore, the eral patterns by age and gravidity, lower socioeconomic status of the non- however, are approximately the same whites, with resulting inability to afford among the two color groups, with the adequate medical care, may mean that ratios consistently higher among the pelvic infections of other than venereal nonwhite women for each age and gra- origin also contribute to the differential vidity category. In both color groups between the two color groups. With age of mother appears to be highly in- a larger number of cases available we fluential in the occurrence of ectopic may be able to test this hypothesis by pregnancy, while women who conceive determining whether the incidence of repeatedly seem to suffer less from this ectopic pregnancy is higher among white aberration. women of poor socioeconomic circum- Previous Fetal Loss-As shown in stances than it is among similar women Table 2, however, earlier fetal loss car- in the more privileged economic group. ries greater risk of ectopic pregnancy. The increase in rate with advancing The ratio rises from 2.4 per 1,000 live age of mother is understandable if one births among women who have never postulates that functional and genetic previously lost a fetus to 9.5 among disturbances are more likely to occur those who have had three or more un- among older women. Yet, the lower successful prior pregnancies. Unfor- incidence among multigravida in each tunately, we do not know their preg- age group requires other explanation. Is it possible that women having func- tional disturbances which result in ecto- Table 2-Ectopic Pregnancies by Num- pic pregnancies during their early ber of Previous Fetal Deaths, New child-bearing history are less likely to York City, 1952-1954 conceive again at all, or that surgical procedures in connection with an initial Number of Pre- Number of Ratio per vious Fetal Ectopic 1,000 ectopic pregnancy, or for conditions that Deaths Pregnancies Live Births would cause ectopic pregnancy, con- tribute to failure to conceive again? Total 1,413 * 2.9 If this is so, some women who have None 1,027 2.4 experienced one ectopic pregnancy One 220 4.8 would be automatically excluded as Two 65 5.8 candidates for a succeeding similar ac- Three or more 55 9.5 cident. Such a situation could account for these * Includes 46 with number of previous fetal deaths findings. not reported. On the other hand, higher incidence FETAL WASTAGE VOL. 46 831

of ectopic pregnancy has been shown to During the triennium, 1952 to 1954, occur among women with history of there were 55,107 presumably sponta- repeated fetal loss. This observation neous fetal deaths reported, of which suggests that fetal loss, perhaps , 41,790, or 76 per cent, had occurred at may be a factor in ectopic pregnancy. less than 20 weeks gestation. Of the Later data in this report reveal that pre- latter, 30,200 were reported among sumably spontaneous abortion is sub- whites and 11,590 among nonwhites. stantially higher among nonwhites. It (Pertinent data are tallied in Table 3). is possible that the higher frequency of Age of Mother, Gravidity, and Color- fetal loss among this group means more The frequency of spontaneous early fetal frequent induced abortion at an early death increases with maternal age from age (reported as spontaneous or not re- 67.2 per 1,000 live births among women ported at all) and that hence the non- under 20 to 193.4 among women 40 whites are more vulnerable to ectopic years old and over (without regard to pregnancies. color). The ratios, moreover, increase steadily with advancing orders of preg- Early Spontaneous Fetal Deaths nancy. Since age and gravidity are closely related, more profit results from Included in the term "spontaneous" their discussion in combination. there may be an undetermined number The figures given in Table 3 appear of abortions which were actually in- difficult to interpret, but the relation- duced, but which were not so identified ships become clear in the Figures 1 and by the attending physician in the case 2 for whites and nonwhites, respectively. report. In effect, what we are dealing Unfortunately, the numbers for non- with here are all pregnancies terminated whites are too small to allow analysis prior to the 20th week of gestation not in the same detail as for the whites. reported as due to therapeutic interven- Among Grava I women the frequency tion, illegal intervention, or ectopic of spontaneous early fetal death rises pregnancy. consistently with age. The rise is rela-

Table 3-Spontaneous Fetal Deaths Under 20 Weeks Gestation, by Age of Mother and by Color and Gravidity, Ratios per 1,000 Live Births, New York City, 1952-1954

White Nonwhite

Gravidity Gravidity

5 or 3 or Age of Mother Total 1 2 3 4 More Total 1 2 More

All ages 72.9* 58.9 54.3 81.9 108.6 132.6 153.5* 106.8 121.1 184.7 Under 15 t f - - - N 15-19 55.1 42.9 74.5 153.5 162.8 370.4 90.0 67.7 101.9 163.8 20-24 58.1 44.7 52.6 87.8 136.4 191.3 145.9 108.1 123.0 183.9 25-29 62.4 57.7 43.2 68.9 95.8 135.6 175.0 128.9 121.6 201.1 30-34 80.0 91.8 54.1 71.7 96.7 124.5 167.9 167.4 122.5 170.9 35-39 111.9 137.2 92.9 102.5 111.0 113.6 168.2 178.1 131.2 169.3 40-44 186.6 251.1 158.3 190.5 176.6 159.1 C185.8 252.5 223.0 169.3 45 and over 384.4 644.4 431.1 491.8 290.9 261.4

* White total includes 151 with age and 1,054 with gravidity not reported; nonwhite total includes 78 with age and 376 with gravidity not reported. f Ratios were not computed where there were fewer than 10 cases. 832 JULY 1956 AMERICAN JOURNAL OF PUBLIC HEALTH

of spontaneous fetal death at advanced age for pregnancies after the first. Among the nonwhites the risk declines consistently among women 40 years and over with more frequent pregnancies, the exact reverse of the situation among t 4X '^' 1 //- the women under 25. But this exact reversal does not occur among the 0 200 whites. In the latter group the risk among Grava II women of 45 years and over is close to, but less than, that for Grava III women. The ratios for women having higher order 0 pregnancies 15-I 20.24 25-29 30-4 6-39 40-44 450 are considerably lower than those for Figure i-Spontaneous Fetal Deaths Grava II's late in the child-bearing Under 20 Weeks Gestation by Age of period. Mother and Gravidity, Whites Only, Some of the New York City, 1952-1954 apparent aberrations in the observed patterns may be due to deficiency in the reporting of previous tively gradual at early ages among the pregnancies, but true relationships be- whites, but very rapid among the non- tween age and gravidity for spontaneous whites throughout the child-bearing age abortions could also be obscured by in- span. For higher gravidity orders the clusion with this group of an unknown patterns vary in the two color groups. number of induced abortions for which The risk for various pregnancy orders the age-gravidity patterns may differ. differs at each end of the age scale. In- The regularity of the patterns, particu- creasingly high ratios at the younger larly among the whites, does seem re- ages for women having second and later markable, however. pregnancies is characteristic of both Already noted is a lower ratio among color groups. Both show smaller risk white women having a second pregnancy than for other orders, but this is true only for specific age groups beginning 300 with age 25. At younger age groups GRAVA I the ratio increases steadily with gravid- W9A II ... GRAIVAIII / ity. Despite the fact that the gross ratios increase consistently with age among the nonwhites, the age-gravidity i200 , -'" pattern of fetal loss is the same as among the whites, although the ratios are uni- formly higher among the nonwhites until

100 age 40 has been reached. These data imply an optimal time for child-bearing and that departures carry with them increased risk of loss of the fetus at an early period of gestation. 0 These "optimum times" appear to be different Figure 2-Spontaneous Fetal Deaths among whites and nonwhites. Under 20 Weeks Gestation by Age of In both groups there is least fetal loss Mother and Gravidity, Nonwhites Only, for first pregnancies when the mother New York City, 1952-1954 is less than 20 years of age, although FETAL WASTAGE VOL. 46 833 the risk is only a little higher at 20-24 the nonwhites as compared with the among the whites. The optimum age slower increase among the whites is for second pregnancies lies between 25 perhaps due to earlier effects of higher and 29 for the whites, but the extra- incidence of disease and poorer nutri- ordinarily and comparatively high ratio tion among the former group. A high of early fetal loss at each age among incidence of uterine tumors may also be the nonwhites makes it appear to differ a factor. little whether the second pregnancy oc- Previous Fetal Deaths-In Table 4 curs at any age within the broad span are presented the ratios of fetal deaths of 20-34. Third and later pregnancies to live births by previous history of among nonwhites are uniformly accom- fetal loss. These data confirm observa- panied by relatively high frequency of tions by many investigators that history fetal loss, but frequent child-bearing at of earlier fetal loss creates greater risk early ages seems to carry an additional of loss in the current pregnancy. Among hazard, as indicated by the high ratios women with no such history the ratio among these women having third or of early fetal loss is 66 per 1,000 live later pregnancies at 20-29 years of age. births (for both color groups combined). Further breakdown by order of preg- But among women who have already nancy is needed among the nonwhites lost one fetus the risk is 2.6 times as to see the extent to which the given great. And for women who have had figures are weighted by higher order four or more unsuccessful pregnancies pregnancies. the risk is nearly 12 times as high as Among the whites, however, where among women without earlier preg- more refined data have been prepared, nancy loss. The same etiologic factors: third pregnancies, as with second preg- environmental, medical, dietary, genetic, nancies, carry least risk to the fetus which led to the first loss, continue to when the mother is between 25 and 29 exist and thus are likely to produce years of age, although the early fetal unfavorable results in later pregnancies. death ratio is not appreciably different These patients are a vulnerable group, among women 30-34 years old. Fourth susceptible to continued fetal loss. It pregnancies should occur, it appears, is regrettable that we do not have in- between 25 and 34 years, also; but least formation as to the type of the earlier fetal loss in fifth and later pregnancies occurs when the mother is 35-39 years old. Table 4Spontaneous Fetal Deaths Un- der 20 Weeks Gestation by Number In summary, for the whites it seems of Previous Fetal Deaths, New desirable that first pregnancies take York City, 1952-1954 place before a woman is 25 years old, second to fourth pregnancies should be Number of Pre- Number of Ratio per spaced in the 10-year interval 25-34, vious Fetal Spontaneous 1,000 while fifth or higher pregnancies carry Deaths Fetal Deaths Live Births least risk to the fetus at maternal ages of 35-39. The desirability of proper Total 41,790* 85.3 spacing of children is a clear implication. None 28,207 66.1 Abnormalities and disease conditions One 7,146 155.8 of the placenta or aging and defective Two 2,406 217.5 genes of the mother may be responsible Three 999 275.2 for the increasing incidence of spontane- Four or more 1,661 765.4 ous abortion with advances in maternal * Includes 1,371 with number of previous fetal deaths age. The sharp rise with age among not reported. 834 JULY 1956 ANIERICAN JOURNAL OF PUBLIC HEALTH fetal death, i.e., whether spontaneous, Better prenatal care, adequate nutrition induced, or ectopic pregnancy. for mothers before and during preg- Cause of Fetal Death-The paucity nancy, and improvement in obstetric of knowledge about early fetal mortality practices would undoubtedly result in is nowhere more clearly illustrated than salvaging many lives and wasted preg- in a discussion of causes. Almost 75 nancies, but much is still unknown. per cent of the spontaneous losses are of All conditions under which fetal loss unknown cause. The largest reported may occur must be properly determined factor is "hemorrhage without mention and identified. of placental condition," but the majority More research is needed in the vari- of these cases were reported simply as ous facets of the problem, including the "bleeding" or "spotting." If these cases physical, environmental, biologic, and are excluded as meaningless, the pro- psychologic. This requires a "team ap- portion of unknown cause then be- proach" and the skills of the obstetrician, comes nearly 96 per cent and causes family physician, public health worker, observed in the mother are reduced to pathologist, statistician, nutritionist, psy- less than 3 per cent. Among the known chologist, psychiatrist, social worker, conditions in the mother, ascribed by geneticist, and biochemist. Preventive the physician as cause of the fetal death, measures must be directed at removing the largest group (with the exclusion the responsible or contributory factors mentioned) is chronic maternal disease, in the large number of fetal deaths. in which fibromyomata uteri is the most We know that there appears to be a frequent condition stated, with infec- distinct relationship between previous tion and toxemia following in order of unfavorable pregnancies and subsequent frequency. fetal or early neonatal loss, indicating Placental and cord conditions account a possible vulnerable group of women. for nearly 40 per cent of the early fetal It is highly important that these indi- deaths of stated cause (again with the viduals be carefully studied and that exclusion noted), more than half spe- appropriate therapeutic measures be in- cifying premature separation of the stituted before they conceive again. placenta. Among the whites slightly Individuals with a previous history more than 3 per cent of this group of of spontaneous abortion or ectopic preg- early fetal deaths were ascribed to con- nancy deserve special attention in well genital malformations and an equal organized and approved preconceptual proportion to erythroblastosis. Report- treatment clinics. Both parents must be ing of cause was notably more deficient studied, since defective genes of either among nonwhites than among whites, parent or psychological and biological but even in the latter only one of 20 disturbances in either or both may be reports had a definitive statement of causative factors and determining in- cause. Under these circumstances the fluences in fetal loss. value of analysis of these reports by cause is highly questionable. Conclusions Comment-Fetal loss is due to many conditions, including ectopic pregnancy Data are presented on fetal loss due and spontaneous abortions, and is a to ectopic pregnancies and spontaneous major public health problem today. The abortion in New York City, 1952-1954. causes are multiple and the solution of Both show a preponderance among the the problem requires concerted multi- nonwhite, the ratio being about three disciplinary action. Further study and times higher than that among the whites. intensive investigation are necessary. The nonwhite spontaneous abortion FETAL WASTAGE VOL. 46 835

ratio is more than double that of the Preconceptual clinics, including the serv- whites. ices of obstetricians, family physician, Ectopic pregnancies increase with psychiatrist, geneticist, statistician, bio- advancing age of the mother. The rate chemist, nutritionist, public health of increase is higher in first pregnan- worker, and social worker, who will cies. Earlier fetal loss carries greater study both parents will help in the pre- risk of subsequent fetal loss, both in vention of fetal loss and in the ultimate ectopic pregnancies and spontaneous saving of many lives. abortions. It appears desirable that first preg- REFERENCES nancies take place before age 25, while 1. Erhardt, Carl L. Reporting of Fetal Deaths in New York City. Pub. Health Rep. 67:12 (Dec.), 1952. fifth pregnancies should not occur be- 2. Baumgartner, Leona, and Erhardt, Carl L. Some fore age 35 among the whites. No firm Observations on the Factors in the Incidence of Pre- mattiritv and Fetal Death. In Chapter X of "Preg- optimal ages for child-bearing could be nancy Wastage." Springfield, Ill.: Thomas, 1953. determined for the nonwhites on the 3. Baumgartner, Leona, et al. The Inadequacy of Rou- tine Reporting of Fetal Deaths. A.J.P.H. 39:12 basis of these data, except that first (Dec.), 1949. pregnancies carry least loss at less than 4. Tietze, Christopher. Therapeutic Abortions in New York City, 1943-1947. Am. J. Obst. & Gynec. 60:1 20 years of age. There is a distinct (July), 1950. lack of reliable information about the 5. Wiehl, Dorothy G. A Summary of Data on Reported Incidence of Abortion. Milbank Mem. Fund Quart. cause of fetal mortality, and basic re- XVI: No. 1 (Jan.), 1938. search is needed on the 6. Stix, Regine K. A Study of Pregnancy Wastage. various causes Ibid. XIII:4 (Oct.), 1935. of pregnancy wastage. 7. Anderson, George W. The Racial Incidence and Sound preventive measures Mortality of Ectopic Pregnancy. Am. J. Obst. & based on Gynec. 61:2 (Feb.), 1951, and Fontanilla, Jose, and reliable and well scrutinized information Anderson, George W. Further Studies on the Racial Incidence and Mortality of Ectopic Pregnancy. Ibid. must be applied prior to conception. 70:2 (Aug.), 1955.

Professional Nursing Examinations The U. S. Civil Service Commission announces a new examination for filling professional nursing positions in the Clinical Center of the National Institutes of Health. Graduate professional nurses and students who are within six months of completing their nursing courses are eligible, although the latter cannot enter on duty until their courses are completed. The examination consists of a review of experience, education, and training. Entrance annual salaries for four different levels of positions range from $3,670 to $5,440. Applications will be accepted until the needs of the service have been met. Executive Secretary, Board of U. S. Civil Service Examiners, National Institutes of Health, Bethesda, Md.