Journal of Perinatology (2013) 33, 817–820 & 2013 Nature America, Inc. All rights reserved 0743-8346/13 www.nature.com/jp

PERINATAL/NEONATAL CASE PRESENTATION Non-immune hydrops fetalis caused by herpes simplex virus type 2 in the setting of recurrent maternal infection

KM Pfister1, MR Schleiss2, RC Reed3 and TN George1

We report a case of non-immune hydrops fetalis (NIHF) caused by herpes simplex virus type 2 (HSV-2) in an infant whose mother had recurrent HSV-2 infection. In spite of prematurity, severe disseminated infection and hydrops, the infant survived and was neurologically intact. HSV-2-induced NIHF is extremely rare, particularly in the setting of recurrent maternal infection, and this case is, to our knowledge, the first report of a surviving infant. HSV-2 should be considered in the differential diagnosis of NIHF and early initiation of empiric acyclovir therapy is recommended in this setting, pending the results of virologic diagnostic tests.

Journal of Perinatology (2013) 33, 817–820; doi:10.1038/jp.2013.68 Keywords: neonatal HSV infection; fetal hydrops; HSV-2 infection; placental infection; acyclovir; torch infection

INTRODUCTION Following transfer to our neonatal intensive care unit from a Non-immune hydrops fetalis (NIHF), which occurs in 1 in 2500 to referring facility on day of life (DOL) 1, examination was remark- 4000 , continues to have a very high able for a severely hydropic-appearing premature infant with a rate, ranging from 50 to 90%.1,2 Cardiac disorders, genetic distended abdomen and an enlarged, firm liver. Skin examination abnormalities, fetal malformations, hematologic disorders and showed diffuse erythema, non-tense bullae on the chest and infections can all lead to NIHF. Among the infectious etiologies of extremities, as well as superficial erosions and hemorrhagic macules NIHF, the most common causes are , toxoplasmosis, and papules. The face and scalp were notably spared (Figure 1). and infection. Herpes simplex virus (HSV) is Initial lab values demonstrated a white blood cell count of only rarely described as a cause of NIHF, typically in the setting of 10.4 Â 109 l À 1, a hemoglobin of 8 g dl À 1, a platelet count of primary maternal infection, and no surviving infants have been 38 Â 109 l À 1, a direct antiglobulin and that were both described in the previous case reports.3–10 We report the case of an negative, an international normalized ratio of 2.87 (reference range, infant with NIHF secondary to intrauterine HSV-type 2 (HSV-2) 0.81 to 1.30), a partial thromboplastin time of 89 s (reference range, infection who survived; notably, this infant was born to a woman 27 to 52 s), a fibrinogen of 198 g dl À 1 (reference range, 200 to with recurrent HSV-2 infection. 420 mg dl À 1) and an aspartate aminotransferase and alanine aminotransferase of 468 (reference range, 20 to 65 U l À 1)and41 (reference range, 0 to 50 U l À 1)UlÀ 1,respectively. CASE The infant required high-frequency oscillatory ventilation for A 24-year-old gravida 1 para 0 woman presented at 27-1/7 weeks’ severe hypoxemic respiratory failure. Initial echocardiogram gestation with preterm labor and spontaneous rupture of showed evidence of pulmonary hypertension with tricuspid membranes for 2 h. Her had been uncomplicated to regurgitation and right-to-left shunting through the patent that point, with normal ultrasounds at 12 and 20 weeks. Prenatal foramen ovale and patent ductus arteriosus, as well as severely labs were unremarkable and she was HIV negative. The patient had depressed biventricular function and a small a history of a primary genital HSV-2 infection approximately 4 years without tamponade. Inhaled nitric oxide (iNO) was initiated, prior to this pregnancy, documented at that time by culture and resulting in a change of the patent ductus arteriosus shunt to type-specific serology (HSV-2 immunoglobulin G positive). She bidirectional status, and subsequent clinical improvement was denied a history of symptomatic recurrent genital HSV infections, noted. Although iNO therapy is not routinely recommended for and she had not received suppressive nucleoside therapy prior to severe respiratory failure and has not been shown to be beneficial or during this pregnancy. At presentation, she was fully dilated, in this setting,11,12 a recently published NIH consensus statement and a hydropic-appearing male infant was spontaneously deli- on iNO therapy for premature infants (http://consensus.nih.gov/ vered several minutes after admission. The infant had no 2010/inofinalstatement.htm) noted that there are clinical spontaneous respiratory effort or movement, with an initial heart situations, including pulmonary hypertension, in which iNO may rate less than 60. Resuscitation included intubation, positive have benefit in preterm infants. Given the clinical presentation pressure ventilation, chest compressions, intravenous epinephrine and the echocardiogram results, we felt that iNO would be and intravenous fluids. Surfactant was administered. Apgar scores beneficial to this infant. Medical therapies included empiric were 4 and 7 at 1 and 5 min, respectively. ampicillin and gentamicin, and pressor support (dopamine, 3 to

1Division of Neonatology, Department of Pediatrics, University of Minnesota Amplatz Children’s Hospital, Minneapolis, MN, USA; 2Center for Infectious and Microbiology Translational Research, Division of Pediatric Infectious Diseases and , Department of Pediatrics, University of Minnesota Amplatz Children’s Hospital, Minneapolis, MN, USA and 3Department of Laboratory Medicine and Pathology, University of Minnesota Amplatz Children’s Hospital, Minneapolis, MN, USA. Correspondence: Dr MR Schleiss, Center for Infectious Diseases and Microbiology Translational Research, Division of Pediatric Infectious Diseases and Immunology, Department of Pediatrics, Minneapolis, MN 55455, USA. E-mail: [email protected] Received 15 April 2013; accepted 8 May 2013 Non-immune hydrops fetalis caused by herpes virus KM Pfister et al 818

Figure 1. Skin findings in intrauterine HSV-2 infection. Multiple bullae and erosions are noted on trunk and extremities. HSV, herpes simplex virus.

20 mcg kg À 1 min À 1, for 3 days with peak dosing at approximately 48 h of life, and dobutamine 3 to 7 mcg kg À 1 min À 1, for 12 h on DOL 2). Corticosteroid therapy for clinical adrenal insufficiency was initiated on DOL 3 (hydrocortisone sodium succinate, 4mgÀ 1 kg À 1 day for 2 days, which was tapered over the follow- ing 5 days). Intravenous immunoglobulin (2 doses of 0.5 g kg À 1 each on DOL 3 and 4), and multiple packed red blood cell, platelets, and fresh frozen plasma transfusions were also administered. Empiric acyclovir therapy was initiated at 8 h of life. Abdominal ultrasound demonstrated diffuse hepatic microcalcifications. Head ultrasound revealed hemorrhage in the left ventricle without ventricular dilation with a small periventricular hemorrhage, as well as right germinal matrix hemorrhage. Diagnostic studies obtained to elucidate the potential etiology of NIHF included parvovirus, toxoplasma and treponema serolo- Figure 2. Herpes simplex viral inclusions in amniotic epithelium gies; blood and urine cytomegalovirus culture and polymerase of fetal membranes. (a) Hematoxylin and eosin (H&E) stain chain reaction (PCR); and blood PCR for HIV and enterovirus, all of showing eosinophilic intranuclear viral inclusion (arrow). (b) H&E which were negative. The total serum immunoglobulin M level stain showing chromatin margination and ground-glass nuclear viral inclusions (arrows). (c) Immunoperoxidase stain for HSV-1 and 2. was 234 mg dl À 1. Blood HSV-2 PCR was positive; in addition, Magnification  400. HSV, herpes simplex virus. HSV-2 was also identified by both conjunctival viral culture and PCR, and from the nasopharynx and rectum by PCR. Ophthalmo- logic exam revealed bilateral herpetic keratitis. A lumbar puncture lesions resolved, although post-inflammatory hypopigmentation was not possible due to clinical instability. Placental pathology was noted. Brain MRI prior to discharge showed moderate atrophy demonstrated viral cytopathic changes in amniotic epithelial cells of the right cerebellar hemisphere and slightly less than expected of the fetal membranes, including chromatin margination and T1 hyperintensity in the corticospinal tracts and posterior limb of eosinophilic intranuclear inclusions. Immunohistochemical stain- the internal capsule on the left. There were no cystic changes. The ing with an HSV-specific monoclonal antibody confirmed the infant was discharged on DOL 114 at 43-2/7 weeks corrected age presence of HSV (Figure 2). There was also evidence of acute and with a plan to continue supplemental low-flow oxygen therapy chronic funisitis and acute , but no villitis. and gastrostomy tube feeds. Oxygen therapy was gradually HSV DNA was no longer detected in blood by DOL 10. weaned and discontinued by 6 weeks post-discharge. The infant was treated with high-dose intravenous acyclovir Upon discontinuation of acyclovir, after 6 months of suppres- (40 mg kg À 1 per day) for 30 days and then transitioned to oral sive therapy, the patient had recurrence of skin lesions. He was acyclovir suppressive therapy (300 mg m À 2 three times per day) again treated with intravenous acyclovir for 48 h, and transitioned for 6 months.13 Topical trifluridine was also administered for to oral suppressive therapy for an additional 3 months, with herpetic keratitis and was continued until this resolved. The skin prompt resolution of his lesions.

Journal of Perinatology (2013), 817 – 820 & 2013 Nature America, Inc. Non-immune hydrops fetalis caused by herpes virus KM Pfister et al 819 Table 1. Previously reported cases of NIHF secondary to HSV infection

Author ref. Live born Death Frank Fluid locations HSV-type Maternal hydrops infection

Im et al.3 Not specified Not specified Yes Not specified HSV-1 Not specified Greene and No Fetal Yes Body wall, pleural, HSV-1 Not specified Wirtz4 Dubois-Lebbe No Fetal Yes Body wall, placenta, HSV-1 Primary et al.5 no autopsy Anderson and Yes – 29-1/7 Withdrawal of care on day of life 2 Yes Body wall, ascites Not Primary Abzug6 weeks specified Beck et al.7 Yes – 25-0/7 Withdrawal of care after poor response to Yes Body wall, ascites HSV-2 Primary weeks resuscitation Barefoot et al.8 Yes – 31-2/7 Did not survive resuscitation No Ascites only HSV-2 Not specified weeks Christie et al.9 Yes – 31-0/7 Died on day of life 4 No Body wall only HSV-2 Uncleara weeks Lee et al.10 Yes – 28 Withdrawal of care on day of life 2 No only HSV-2 Probable weeks primary Abbreviations: HSV, herpes simplex virus; NIHF, non-immune hydrops fetalis. aMaternal HSV titer 1:16, nonspecific for immunoglobulin G or immunoglobulin M.

Neurodevelopmental outcome at 12 months corrected gesta- the setting of documented recurrent maternal infection was tional age was performed using the Bayley Scales of Infant unexpected, the infant survived and had a good long-term Development —Third Edition, a comprehensive measure of neurologic outcome. Although a large proportion of maternal general intellectual ability that provides separate scores for immunoglobulin G transfer to the during the cognitive, language and motor domains. This evaluation demon- third trimester of pregnancy, maternal immunoglobulin G can strated that his cognitive function was within the average range, reach the fetus as early as 17 weeks’ gestation.20 With a presumed and that he had only minimal-to-mild motor and language delays. recurrent HSV-2 infection in pregnancy leading to the baby’s His cognitive composite score was 90 (±1 s.d. ¼ 85 to 115) with presentation, the mother may have high levels of HSV-2-specific an age equivalent of 10 months. The overall language composite antibodies that protected this infant despite his premature birth. score was 79 (±1 s.d. ¼ 85 to 115). Finally, he performed at the 10- The absence of data on maternal and neonatal HSV-2 antibody month age equivalency on the measure of fine motor ability. His levels in this case makes this theory speculative, but may help gross motor skills, including locomotion, coordination, balance explain the good outcome. In addition, the prompt administration and motor planning, were within the 9-month age equivalency of high-dose acyclovir (started within 8 h of birth) likely with an overall motor composite score of 82 (± 1 s.d. ¼ 85 to 115). contributed to this infant’s survival. The role of cell-mediated immunity in resolution of neonatal HSV infection is poorly understood,21 as is the role of the interferon response.22 Infants DISCUSSION with genetic polymorphisms in the toll-like receptor 3 gene are NIHF due to congenital HSV infection has only rarely been predisposed to more severe neonatal HSV infections,23 but reported.3–10 Our review of the literature identified only five cases unfortunately TLR3 polymorphism analysis was not available for of frank hydrops caused by HSV (Table 1). This case, to our the infant described in this case. knowledge, is the first reported case of survival of an infant with In conclusion, although a rare cause of NIHF, HSV infection HSV-induced NIHF. Moreover, our case is unique insofar as it should always be considered in the differential diagnosis, even in occurred in the context of recurrent maternal HSV-2 infection. the setting of a maternal history of recurrent HSV. Empiric early Most neonatal HSV infections result from exposure to the virus treatment with acyclovir should be initiated in a neonate with in the genital tract during delivery.14 It is estimated that NIHF of unknown etiology while awaiting diagnostic studies, as intrauterine infection occurs in only 4 to 5% of neonatal HSV this may significantly improve the chance of survival. cases.15,16 Additionally, in the largest and most recent review on intrauterine HSV infection, only 6 of 64 reported infants presented with hydrops.15 Intrauterine infection can occur as a consequence CONFLICT OF INTEREST of transplacental or ascending infection. Evidence of HSV found in The authors declare no conflict of interest. the amniotic epithelium of the fetal membranes without villitis (Figure 2) suggests ascending and not transplacental transmission in this case. Although ascending infection without rupture of ACKNOWLEDGEMENTS membranes is uncommon, this mode of transmission has been We acknowledge the useful discussions with Planned Parenthood of St Paul, reported previously with placental pathology similar to the Minnesota. Funding support from HD044864 and HD068229 is acknowledged. present case.16–19 In the aforementioned cases, primary genital HSV infection was proven or presumed based on maternal history. Interestingly, the mother in our case had long-established , REFERENCES with documented genital HSV-2 infection 4 years prior to this 1 Murphy JH. Nonimmune hydrops fetalis. Neoreviews 2004; 5: e5–e15. pregnancy. The vast majority of cases of neonatal HSV infection 2 Wolf RB, Moore TR. Amniotic fluid and nonimmune hydrops fetalis. In Martin RJ, occur in the setting of primary maternal infection during Fanaroff AA, Walsh MC (eds) Fanaroff and Martin’s Neonatal-Perinatal Medicine. Eighth Edition Mosby Elsevier: Philadelphia, PA, USA, 2006, pp 420–422. pregnancy. The reduced risk of neonatal HSV in the setting of 3 Im SS, Rizos N, Joutsi P, Shime J, Benzie RJ. Nonimunologic hydrops fetalis. recurrent maternal infection is believed to be related to Am J Obstet Gynecol 1984; 148(5): 566–569. transplacental transfer of antibodies mitigating the transmission 4 Green D, Watson W, Wirtz PS. Non-immune hydrops associated with congenital 14 of infection. Although congenital HSV and NIHF in our patient in herpes simplex infection. SDJMed1993; 46: 219–220.

& 2013 Nature America, Inc. Journal of Perinatology (2013), 817 – 820 Non-immune hydrops fetalis caused by herpes virus KM Pfister et al 820 5 Dubois-Lebe C, Houfflin-Debarge V, Dewilde A, Devisme L, Subtil D. 14 Corey L, Wald A. Maternal and virus infections. N Engl J Nonimmune hydrops fetalis due to herpes simplex virus type 1. Prenat Diagn 2007; Med 2009; 361(14): 1376–1385. 27: 188–189. 15 Marquez L, Levy MM, Munoz FL, Palazzi DL. A report of three cases and review 6 Anderson MS, Abzug M. Hydrops fetalis: an unusual presentation of intrauterine of intrauterine herpes simplex virus infection. Pediatr Infect Dis J 2011; 30(2): herpes simplex virus infection. Pediatr Infect Dis J 1999; 18: 837–839. 153–157. 7 Beck R, Park T, Farrington P, Steigman CK, Sennesh JD. Congenital disseminated 16 Chaterjee A, Chartrand SA, Harrison CJ, Felty-Duckwort A, Bewtra C. Severe herpes simplex virus type II infection in a premature infant. Am J Perinatol 1987; 4: intrauterine herpes simplex disease with placentitis in a newborn of a mother 334–338. with recurrent genital infection at delivery. J Perinatol 21: 559–564. 8 Barefoot K, Little G, Ornvold K. Fetal demise due to herpes simplex virus: an 17 Von Herzen JL, Benirschke K. Unexpected disseminated herpes simplex infection illustrated case report. J Perinatol 2002; 22: 86–88. in a newborn. Obstet Gynecol 1977; 50(6): 728–730. 9 Christie J, Rakusan T, Martinez M, Lucia HL, Rajaraman S, Edwards SB et al. 18 Hain J, Doshi N, Harger JH. Ascending transcervical herpes simplex infection with Hydranencephaly caused by congenital infection with herpes simplex virus. intact fetal membranes. Obstet Gynecol 1980; 56(1): 106–109. Pediatr Infect Dis J 1986; 5: 473–478. 19 Hyde SR, Giacoia GP. Congenital herpes infection: placental and 10 Lee A, Bar-Zeev N, Walker S, Permezel M. In utero herpes simplex encephalitis. findings. Obstet Gynecol 1993; 5(2): 852–855. Obstet Gynecol 2003; 102: 1197–1199. 20 Saji F, Samejima Y, Kamiura S, Koyama M. Dynamics of immunoglobulins at the 11 Donohue PK, Gilmore MM, Cristofalo E, Wilson RF, Weiner JZ, Lau BD et al. feto-maternal interface. Rev Reprod 1999; 4(2): 81–89. Inhaled nitric oxide in preterm infants: a systematic review. Pediatrics 2011; 21 Muller WJ, Jones CA, Koelle DM. Immunobiology of herpes simplex virus and 127(2): e414–e422. cytomegalovirus infections of the fetus and newborn. Curr Immunol Rev 2010; 12 Askie LM, Ballard RA, Cutter GR, Dani C, Elbourne D, Field D et al. Inhaled nitric 6(1): 38–55. oxide in preterm infants: an individual-patient data meta-analysis of randomized 22 Han JY, Schleiss MR. It is time to examine the role of host cytokine response in trials. Pediatrics 2011; 128(4): 729–739. neonatal herpes simplex virus infection. Future Virology 2011; 6(6): 679–681. 13 Kimberlin DW, Whitley RJ, Wan W, Powell DA, Storch G, Ahmed A et al. Oral 23 Zhang SY, Jouanguy E, Ugolini S, Smahi A, Elain G, Romero P et al. TLR3 acyclovir suppression and neurodevelopment after neonatal herpes. N Engl J Med deficiency in patients with herpes simplex encephalitis. Science 2007; 317(5844): 2011; 365(14): 1284–1292. 1522–1527.

Journal of Perinatology (2013), 817 – 820 & 2013 Nature America, Inc.