Rapid Development ofNephrotic Syndrome,Hypertension, and HemolyticAnemia Early in Pregnancyin Patients With IDDM

Ruth S. Weinstock, MD, PhDRichard T. Kopecky, MDDavid B. Jones, MDShiraz Sunderji, MD

In recent years, the prognosis for a successfulpregnancy has greatly improved for women with insulin-dependent diabetes mellitus (IDDM) who are undergood glycemic control and free of complications such asvascular disease and nephropathy. We report the rapiddevelopment of severe nephrotic syndrome, malignanthypertension, and microangiopathic hemolytic anemiaduring the first trimester of pregnancy in a 29-yr-oldwoman with IDDM of 18 yr duration. Our patienthad no pregestational history of retinopathy orhypertension and only minimal proteinuria. Significantimprovement in blood glucose levels had been achievedover the 6 mo before conception. Kidney biopsyperformed before the termination of pregnancy at 10 wkgestation revealed diabetic nephropathy. No otheretiology for her renal disease could be found. Anarteriole was noted to have entrapped red blood cellfragments and platelet thrombi, revealing the probablesource of her hemolytic process. By 8 wk postpartum,her nephrotic syndrome and hemolysis had completelyresolved. At 3 mo postgestation, the patient'shypertension was still present but less severe. Herserum creatinine has continued to decrease towardnormal. This is the first report of a woman with IDDM inWhite's classification C who developed a toxemia-likesyndrome during the first trimester of pregnancy,attributable to the underlying diabetic state. DiabetesCare 11:416-21, 1988

From the Departments of Medicine, Pathology, and Obstetrics and Gynecol-ogy, the State University of New York Health Science Center, and the Depart-ment of Medicine, Veterans Administration Medical Center, Syracuse, NewYork.

Address correspondence and reprints requests to Ruth S. Weinstock, MD,PhD, Department of Medicine, SUNY Health Science Center, 750 E. AdamsStreet, Syracuse, NY 13210.

Few studies carefully document the effects of preg-nancy on diabetic renal and ocular disease. Re-cent reports suggest that diabetic patients withoutsignificant pregestational renal disease or hyper-

tension do not develop renal-function deterioration dur-ing pregnancy (1). We describe a patient with White'sclassification C diabetes who rapidly developed symp-tomatic diabetic nephropathy complicated by the ne-phrotic syndrome, malignant hypertension, and micro-angiopathic hemolytic anemia during the first trimesterof pregnancy.

CASE REPORT

A 29-yr-old woman was hospitalized because of hyper-tension and the nephrotic syndrome during the firsttrimester of pregnancy. The patient had a history ofinsulin-dependent diabetes mellitus (IDDM) for 18 yr.There was no history of cardiac disease, hypertension,peripheral vascular disease, or neuropathy. She smokedhalf a pack of cigarettes each day and had never usedoral contraceptives. A single previous pregnancy, 6 yrearlier, was not associated with hypertension or renaldisease. During that gestation, glycemic control wasmonitored by urine testing only. A 3097-g baby withMobius syndrome was delivered by cesarean section at37 wk. After the baby was delivered, the patient's dia-betes mellitus was not well controlled. Glycosylatedhemoglobin level 7 mo before the second conceptionwas 16.9% (normal, 5.5-8.5%). Six months beforeconception, she received extensive diabetes teachingand began self-monitoring of blood glucose. Her bloodpressure, hematological, and renal-function parametersat that time and throughout her subsequent clinical

416 DIABETES CARE, VOL. 11, NO. 5, MAY 1988

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022

R.S. WEINSTOCK AND ASSOCIATES

course are shown in Fig. 1. Four months before con-ception, an ophthalmological consultant detected no di-abetic retinopathy. One month later, the glycosylatedhemoglobin was 11.8% (normal, 5.5-9.0%). Begin-ning 3 wk before conception, the patient's insulin reg-imen was intensified from two to three injections perday.

She was seen for the first time at the State Universityof New York (SUNY) Health Science Center during the7th wk of pregnancy. She noted a sore throat and a3.6-kg weight gain since her last menstrual period. Thepatient's height was 169.5 cm, weight 69 kg, and bloodpressure 150/80 mmHg. She was afebrile. The pharynxwas normal, no retinopathy was seen, and 2+ pittingedema extended bilaterally to the midcalf. Laboratoryevaluation revealed a normochromic normocytic ane-mia, 135 mM sodium, 4.4 mM potassium, 111 mMchloride, 25 mM bicarbonate, and 106 mg/dl glucose.Liver-function studies were normal. Glycosylated hemo-globin was 6.3% (normal, 3.4-6.1%). Urinalysis re-vealed specific gravity 1.020, pH 5, protein 4 + ,glucose negative, hemoglobin trace, with two whiteblood cells and eight red blood cells per high-powerfield. Urine and throat cultures were negative. Self-mon-

itoring showed the following blood glucose ranges: fast-ing, 83-163 mg/dl; prelunch, 83-145 mg/dl; pre-dinner, 122-210 mg/dl; and bedtime 128-180 mg/dl.The patient's insulin regimen and diet were extensivelymodified, and she began doing fingerstick blood glucosedeterminations seven times per day. One week later,she had fasting blood glucose values <100 mg/dl andpostprandial blood glucose values <140 mg/dl; how-ever, increasing peripheral edema and proteinuria werenoted.

In the 9th wk of pregnancy, the patient noted blurredvision despite continued excellent glycemic control.The blood pressure was 160-170/90 mmHg, and fun-duscopic examination revealed early papilledema, per-imacular and perifoveal edema, cotton-wool spots, andimpaired visual acuity. She was hospitalized, placed onstrict bed rest, and treated with methyldopa and sodiumrestriction. With this regimen, the blood pressure fell to140/85 mmHg; however, peripheral edema increasedand became generalized. On the 2nd hospital day, anophthalmological consultant detected macular edemawith soft exudates, mild arteriolar narrowing, andscattered dot-and-blot hemorrhages consistent withbackground diabetic retinopathy and hypertension.

BO I

2 O

50-

4 0 -

30-

20-

10-

I

;atin

ine

(m

g/>

o

2.0-

1.5-

1.0-

-140

40 -i

3 0 -

D

3

I o2i) I- PIU

I)

3

3

>. ccs S.£ o

20170150130

1 109 07 0

-c 6A

Pre-Conception(months)

Post-Gestation (weeks)

FIG. 1. Clinical course of 29-yr-old woman with insulin-dependent diabetes mellitus. Pregestational, gestational, andpostgestational data. PRBC, transfusion of packed red blood cells.

DIABETES CARE, VOL. 11, NO. 5, MAY 1988 417

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022

NEPHROTIC SYNDROME IN PREGNANT WOMAN WITH IDDM

FIG. 2. Photomicrograph shows a glomerulus with increased mesangial cells and matrix (G, left) and afferent arteriolewith fibrinoid necrosis (A, right) exhibiting a thrombus containing red blood cell fragments (arrows). Hematoxylin andeosin; X1250.

Laboratory evaluation revealed 2.9 g/dl albumin, 5.4mg/dl uric acid, 490 U/L lactate dehydrogenase (nor-mal, 100-225 U/L), and worsening anemia and throm-bocytopenia. The peripheral blood smear showed manyfragmented erythrocytes, suggestive of microangio-pathic hemolytic anemia. Direct Coombs' test was neg-ative. The reticulocyte count was 13.6%. Assays foranti-DNA antibodies and anti-nuclear antibodies werenegative. Levels of C3, C4, and anti-streptolysin O andprothrombin time, thrombin time, partial thromboplas-tin time, fibrinogen, and template bleeding time werenormal. Fibrin split products were present at a concen-tration of 10-40 |xg/ml. A sonographic study revealeda single intrauterine fetus of 9-wk size and normal ma-ternal kidneys. Blood, throat, and urine cultures werenegative. A urine collection showed a total protein ex-cretion of 6.50 g/day. On the 3rd hospital day, the ureanitrogen and creatinine rose to 42 and 1.9 mg/dl, re-spectively. The serum albumin was 1.8 g/dl, and thelactate dehydrogenase was 582 U/L. The patient re-

ceived three units of packed red blood cells. A percu-taneous renal biopsy was performed on the 4th hospitalday.

Examination of the renal biopsy specimen by lightmicroscopy revealed 13 glomeruli with thickened cap-illary basement membranes, increased mesangial ma-trix, and mesangial cell proliferation characteristic ofmoderate diffuse diabetic glomerulosclerosis. Hyalinearteriosclerosis and hyaline cap lesions were also pres-ent. One glomerulus showed evidence of fibrinoid ne-crosis of an afferent arteriole with entrapped erythrocytefragments and platelet thrombi (Fig. 2). There was noglomerular endotheliosis. Immunofluorescent micros-copy revealed small amounts of IgG, C3, and C1q alongthe capillary wall without any fibrin deposition. Thesefindings were considered nonspecific. Electron micros-copy revealed diffuse glomerular basement membrane-thickening and increased mesangial cells and matrixcharacteristic of diffuse mild diabetic glomerulosclerosis(Fig. 3). No lesions diagnostic of malignant nephro-

418 DIABETES CARE, VOL. 11, NO. 5, MAY 1988

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022

R.S. WLINSTOCK AND ASSOCIATES

sclerosis, hemolytic-uremic syndrome, or preeclampsiawere seen.

On the 5th hospital day, mild pulmonary edema de-veloped. In view of the patient's deteriorating clinicalstatus with decreasing renal function, severe hemolyticanemia, and hypertension, the pregnancy was termi-nated. Pathological examination of decidual fragmentsrevealed no vascular changes of preeclampsia. Afterpregnancy termination, there was gradual improvementin the microangiopathic hemolysis, with increases in thehematocrit and platelet count that were associated withresolution of edema and decreasing urinary protein ex-cretion. The patient remained hypertensive, however,requiring captopril (200 mg/day) and furosemide (20mg/day) for blood pressure control. By the 4th wk post-partum, renal function and proteinuria had improvedbut not to prepartum levels. At 8 wk postpartum, theurinary protein excretion was within normal limits at0.08 g/day. Renal function has continued to slowly im-prove, with serum creatinine falling further to 1.2 mg/dl by 3 mo postgestation. Her antihypertensive medi-cation requirements have also been slowly decreasing.

DISCUSSION

Pregnant women with diabetic nephropathy be-fore conception (White's classification F) are atincreased risk for the development of maternal,fetal, and neonatal morbidity and mortality (2—

5). Increasing proteinuria (6) and preeclampsia (7) occurfrequently in these women, typically during the secondhalf of pregnancy. However, we are not aware of anyreport of a patient with such limited pregestational evi-dence of diabetic renal disease who developed a ful-minant toxemia-like course during the first trimester ofpregnancy as described in this case. Six months beforeconception, this patient had a urinary protein excretionof only 180 mg/day, normal renal function, and no evi-dence of diabetic retinopathy. Although this minimalproteinuria may have represented the microalbuminuriaof very early diabetic nephropathy (8) (possibly exac-erbated by poor metabolic control at that time; 9), itwould not have indicated an increased risk for the preg-nancy according to current criteria (3). Nevertheless,

i*Jr

FIG. 3. Electron micrograph shows glomerular capillaries (C) with thick basement membranes (arrow) and thickenedmesangial matrix (M) characteristic of diffuse diabetic glomerulosclerosis; X4800.

DIABETES CARE, VOL. 11, NO. 5, MAY 1988 419

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022

NEPHROTIC SYNDROME IN PREGNANT WOMAN WITH IDDM

she rapidly developed severe nephrotic syndrome, hy-pertension, and microangiopathic hemolytic anemiaearly in pregnancy.

Several disorders can produce hypertension, renaldisease, and microangiopathy during pregnancy (10),including systemic lupus erythematosus (SLE), dissemi-nated intravascular coagulation (DIC), thromboticthrombocytopenic purpura (TTP), the hemolytic-uremicsyndrome (HUS), and preeclampsia. However, each ofthese diagnoses can, in our opinion, be reasonably ex-cluded in this case. The absence of serological markers,hypocomplementemia, or other clinical manifestationsof collagen vascular disease makes SLE an unlikely di-agnosis. DIC is excluded by normal coagulation studies.Several of the typical manifestations of TTP were notapparent in our patient, including fever, central nervoussystem involvement, and purpura, all of which occur in>90% of cases (11). Moreover, renal involvement inTTP is rarely associated with the nephrotic syndrome orsevere hypertension (12). HUS is also an unlikely di-agnosis for several reasons. Although adult HUS candevelop in association with infections, oral contracep-tive use, pregnancy-related complications (e.g., septicabortion, placental abruption), and in the postpartumperiod after normal gestations, it rarely occurs in thefirst trimester of an otherwise normal pregnancy (12,13).Renal involvement in HUS is typically of acute onset,with severe oligoanuric renal failure; and unlike thiscase, HUS does not produce the nephrotic syndrome(12). In addition, localized intravascular coagulationusually prolongs the thrombin time in HUS (14). Al-though preeclampsia can produce most of the abnor-malities manifested by our patient (15), the gestationalage of her pregnancy at presentation (9 wk) excludespreeclampsia as a tenable diagnosis (16). Occasionalcases of preeclampsia occurring before the 20th wk ofgestation are associated with trophoblastic disease (17),for which there was no pathological evidence in thiscase. Furthermore, the typical glomerular abnormalitiesof severe preeclampsia were totally absent from therenal biopsy specimen (18). The identification of asingle blood vessel undergoing fibrinoid necrosis withassociated erythrocyte fragments and platelet thrombiis not a specific pathological finding and can be seenin any of the aforementioned diseases or as a result ofmalignant hypertension. Thus, none of these diagnosessatisfactorily explains all of this patient's clinical course.

We believe that rapid acceleration of underlying oc-cult diabetic nephropathy probably produced this pa-tient's illness. During the early stages of pregnancy, sheprogressed from minimal proteinuria to a florid ne-phrotic syndrome that was due entirely to diabetic glo-merulosclerosis as evidenced by the renal biopsy results.It is possible that the renal hemodynamic alterations as-sociated with pregnancy (18), which are similar to thoseimplicated in the initiation and progression of diabeticglomerular injury (19), triggered this response. This wasfollowed by the appearance of hypertension, a common

occurrence in diabetic patients when nephropathy be-comes clinically overt (20), that rapidly evolved into amalignant phase. Microangiopathic hemolytic anemiaand thrombocytopenia then resulted from hypertensivevascular injury. The importance of hypertension per sein the pathogenesis of the microangiopathy is under-scored by the fact that termination of the pregnancy didnot, in itself, correct the patient's anemia or thrombo-cytopenia. The microangiopathic process did not beginto resolve until several days postpartum, when the pa-tient's blood pressure was controlled and returned tonear-normal levels (Fig. 1).

It is generally believed that the rate of progression ofdiabetic nephropathy is not adversely influenced bypregnancy (1,3,21). In this case, the urinary protein ex-cretion returned to normal after termination of the preg-nancy, and the serum creatinine is slowly returningtoward the patient's baseline pregestational levels. Be-cause renal function can deteriorate quickly in the nat-ural course of some patients with diabetic nephropathy,and because the duration of follow-up of this patient hasbeen relatively brief, it is difficult to determine whetherrenal function has been lost during gestation andwhether it might otherwise have been preserved. Shealso remains hypertensive despite the complete resolu-tion of edema, but because of the short period of follow-up, it is unknown whether the hypertension will be per-manent.

Recent studies have demonstrated maternal and peri-natal morbidity and mortality rates that approach thoseof the general population in diabetic women withoutvascular complications (22,23). The maintenance ofnormal blood glucose profiles during pregnancy appearsto be critical for this success (1,22,23). In this report,the patient had many years of poor glycemic control,with significant metabolic improvement beginning sev-eral months before conception. Note that there are sev-eral reports of acute worsening of retinopathy afterinstituting strict glycemic control in IDDM patients (24-26). However, improvement in retinopathy and nor-malization of glomerular filtration rates have also beenreported (26,27). Whether the normalization of bloodglucose levels on the background of a predisposed retinaand preclinical diabetic nephropathy contributed to theacute clinical deterioration of our patient under thestress of pregnancy is unknown.

This case demonstrates that, although the prognosisfor diabetic pregnancies has substantially improved overrecent years, these women need to be followed carefullyand cautiously. The pregestational assessment of reti-nopathy and nephropathy are essential but may notcompletely reflect maternal and fetal prognosis.

REFERENCES

1. Peterson CM, Jovanovic L: Natural history of the diabeticrenal-retinal syndrome during pregnancy. In Diabetic

420 DIABETES CARE, VOL. 11, NO. 5, MAY 1988

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022

R.S. WEINSTOCK AND ASSOCIATES

10.

11

12.

13.

14.

Renal-Retinal Syndrome. Vol 3. Friedman EA, L'Esper-ance FA Jr, Eds. New York, Grune & Stratton, 1986, p.471-80White P: Pregnancy and diabetes: medical aspects. MedClin N Am 49:1015-24, 1965 15.Kitzmiller JL, Brown ER, Phillippe M, Stark AR, Acker D,Kaldany A, Singh S, Hare JW: Diabetic nephropathy and 16.perinatal outcome. Am j Obstet Cynecol 141:741-51,1981 17.Main EK, Main DM, Gabbe SG: Factors predictingperinatal outcome in pregnancies complicated bydiabetic nephropathy (class F) (Abstract). Diabetes 18.33 (Suppl.1):201A, 1984Rodman HM: Diabetic nephropathy: impact on the prog-nosis of the pregnant diabetic woman. In The Diabetic 19.Pregnancy. A Perinatal Perspective. Merkatz IR, AdamPAJ, Eds. New York, Grune & Stratton, 1979, p. 57-71Hayslett JP, Reece EA: Effect of diabetic nephropathy on 20.pregnancy. Am I Kidney Dis 9:344-49, 1987Lindheimer MD, Katz Al: Renal disease and pregnancy. 21.Perspect Nephrol Hypertens 3:237-54, 1976Parving HH, Noer I, Deckert T, Evrin PE, Nielsen SL,Lyngsoe J, Mogensen CE, Rorth M, Svendsen PA, Trap- 22.Jensen J, Lassen NA: The effect of metabolic regulationon microvascular permeability to small and large mole-cules in short-term juvenile diabetes. Diabetologia12:161-66, 1976 23.Viberti GC, Pickup JC, Jarrett RJ, Keen H: Effect ofcontrol of blood glucose on urinary excretion of albuminand B2-microglobulin in insulin-dependent diabetes. 24.N Engl) Med 300:638-41, 1979Weiner CP: Thrombotic microangiopathy in pregnancyand the postpartum period. Semin Hematol 24:119-29,1987 25.Amorosi EL, Ultmann JE: Thrombotic thrombocytopenicpurpura: report of 16 cases and review of the literature.Medicine 45:139-59, 1966Humphreys MH, Alfrey AC: Vascular diseases of the kid-ney. In The Kidney. 3rd ed. Brenner BM, Recter FC, Eds. 26.Philadelphia, PA, Saunders, 1986, p. 1175-220Ponticelli C, Rivolta E, Imbasciati E, Rossi E, Manucci 27.PM: Hemolytic uremic syndrome in adults. Arch InternMed 140:353-57, 1980Conte J, Deisol J, Mignon-Conte M, TonThat H, Sue JM:

Acute renal failure and intravascular coagulation. In Ad-vances in Nephrology. Vol. 3. Hamburger J, Crosner J,Maxwell MH, Eds. Chicago, IL, Year Book, 1974, p. 197-240Weiner CP: The clinical spectrum of pre-eclampsia. AmI Kidney Dis 9:312-16, 1987Hughes EC: Obstetric-Gynecologic Terminology. Phila-delphia, PA, Davis, 1972, p. 422Pritchard JA, MacDonald PC, Gant NF (Eds.): WilliamsObstetrics. 17th ed. Norwalk, CT, Appleton-Century-Crofts, 1985, p. 448Lindhemier MD, Katz Al: The kidney in pregnancy. InThe Kidney. 3rd ed. Brenner BM, Recter FC, Eds. Phila-delphia, PA, Saunders, 1986, p. 1253-95Zatz R, Brenner BM: Pathogenesis of diabetic microan-giopathy: the hemodynamic view. Am J Med 80:443-53,1986Drury PL: Diabetes and arterial hypertension. Diabeto-logia 24:1-9, 1983Sims EAH: Serial studies of renal function in pregnancycomplicated by diabetes mellitus. Diabetes 10:190—97,1961Jovanovic L, Druzin M, Peterson CM: Effect of eugly-cemia on the outcome of pregnancy in insulin-dependentdiabetic women as compared with normal control sub-jects. Am J Med 71:921-27, 1981Freinkel N, Dooley SL, Metzger BE: Care of the pregnantwoman with insulin-dependent diabetes mellitus. N EnglI Med 313:96-101, 1985Lauritzen T, Larsen H-W, Frost-Larsen K, Deckert T, StenoStudy Group: Effect of 1 year of near-normal blood glu-cose levels on retinopathy in insulin-dependent diabetics.Lancet 1:200-204, 1983Jovanovic L, Chang S, Peterson CM: Medical manage-ment of the diabetic patient with retinopathy. In Ad-vanced Techniques in Ocular Surgery. Jacobiec F,Sigelman J, Eds. Philadelphia, PA, Saunders, 1984, p.422-32Raskin P, Rosenstock J: Blood glucose control and dia-betic complications. Ann Intern Med 105:254-63, 1986Wiseman MJ, Saunders AJ, Keen H, Viberti G: Effect ofblood glucose control on increased glomerular filtrationrate and kidney size in insulin-dependent diabetes. N Englj Med 312:617-21, 1985

DIABETES CARE, VOL. 11, NO. 5, MAY 1988 421

Dow

nloaded from http://diabetesjournals.org/care/article-pdf/11/5/416/438400/11-5-416.pdf by guest on 18 January 2022