 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 105 No. 2 February 2000, pp. 438-444
 |
ABSTRACT |
|---|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
We describe a term infant with congenital cutaneous candidiasis (CCC), and review all cases in the English literature that reported birth weight and outcome. Presence of an intrauterine foreign body was a predisposing factor for development of CCC and subsequent preterm birth. The most common presentation of CCC in neonates weighing >1000 g was a generalized eruption of erythematous macules, papules, and/or pustules that sometimes evolved to include vesicles and bullae. Extremely low birth weight, premature neonates weighing <1000 g most often presented with a widespread desquamating and/or erosive dermatitis (10 of 15 [67%]), and were at greater risk for systemic infection with Candida spp (10 of 15 [67%]) and death (6 of 15 [40%] than those weighing >1000 g (5 of 48 [10%]; 4 of 48 [8%], respectively). Systemic antifungal therapy is recommended for neonates with burn-like dermatitis attributable to Candida spp, or positive blood, urine, and/or cerebrospinal fluid cultures. Systemic treatment also should be considered for all infants with CCC who have respiratory distress in the immediate neonatal period and/or laboratory signs of sepsis such as an elevated leukocyte count with an increase in immature forms or persistent hyperglycemia and glycosuria.
Key words: skin infection, rash, chorioamnionitis, preterm infant, epidermal barrier.
Congenital cutaneous candidiasis (CCC) presents
within the first 6 days of life as a skin eruption caused by
Candida spp. Although candidal infections are common in the
neonatal period, <100 cases of CCC have been reported in the English
literature since candidal chorioamnionitis was first
described.1 Prenatally acquired candidal infections can
produce a spectrum of disease ranging from a diffuse skin eruption in
the absence of systemic illness to severe systemic disease with or
without cutaneous involvement and resulting in fetal demise or early
neonatal death. In this report, we describe a full-term neonate with
CCC and discuss risk factors, clinical features, diagnosis,
pathogenesis, and management of CCC based on a review of the reported
cases.
A full-term, 3705-g female was born by spontaneous vaginal
delivery to a 31-year-old gravida 4 para 3, rubella immune, healthy woman who tested negative to the Venereal Disease Research
Laboratory test. The mother had received a 7-day course of
erythromycin for an uncomplicated urinary tract infection 3 weeks
before delivery. Apgar scores were 4 and 7 at 1 and 5 minutes,
respectively.
On physical examination, a well-developed, vigorous, afebrile female
was grunting and in mild respiratory distress with nasal flaring and
tachypnea (90 to 100 breaths per min). Breath sounds were equal and
clear bilaterally. On her skin were diffusely distributed, 2- to 4-mm,
erythematous macules, papules, and pustules overlying areas of
confluent macular erythema on her face, trunk, and extremities (Fig
1). The palms, soles, and mucous
membranes were spared. The remainder of the physical examination was
normal.
CASE REPORT
View larger version (107K):
[in a new window]
Fig. 1.
Full-term healthy infant with congenital cutaneous candidiasis,
presenting with generalized erythematous papules and pustules.
Laboratory examination showed a total leukocyte count of 29 500/mm3 with 59% polymorphonuclear leukocytes, 6% band forms, 1% myelocytes, 3% metamyelocytes, 7% lymphocytes, 21% monocytes, and 3% eosinophils. Chest radiograph was normal, and her respiratory distress was most consistent with transient tachypnea of the newborn. Cultures of the blood and urine were negative for bacteria and fungi, and Gram stain of the contents from a pustule did not show bacteria. A potassium hydroxide preparation from a pustule showed pseudohyphae. Culture taken from an unroofed pustule grew Candida albicans.
The infant remained vigorous and fed well. No topical or systemic antifungal therapy was given. The superficial pustules ruptured easily and resolved over 2 to 3 days. She was discharged from the hospital on the fourth day of life, and had no sequelae. Diffuse superficial desquamation of the involved skin occurred after approximately 1 week.
| |
METHODS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
To assist in the development of management guidelines for neonates with CCC, we surveyed the English literature to identify liveborn neonates who presented within the first 6 days of life with a cutaneous eruption attributable to Candida spp. Table 1 summarizes the characteristics of those neonates with CCC for whom birth weight and outcome were reported.2-37 Systemic infection was defined as having a positive blood, urine, and/or cerebrospinal fluid culture for Candida spp; or demonstration of Candida in histopathologic or culture specimens obtained at autopsy.
|
| |
RESULTS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
Risk Factors
Nearly half of neonates (7 of 15 [47%]) with CCC and weighing <1000 g were born to mothers who had an intrauterine foreign body in place, either a cervical cerclage or an intrauterine device (IUD) (Table 1). Presence of an intrauterine foreign body was not associated with CCC in neonates weighing >1000 g (1 of 48 [2%], suggesting that the foreign body tends to predispose to the development of CCC and subsequent preterm birth early, during the first 2 trimesters of gestation. Additional reports of chorioamnionitis and overwhelming systemic infection with Candida spp, leading to fetal demise in women with a cerclage or IUD, further supports this contention.38-47 There was no apparent association between systemic administration of antibiotics (2 of 63 patients) or corticosteroids (4 of 63 total patients, 1 of 15 neonates weighing <1000 g) within 1 month of development of CCC (data not shown).
Clinical Presentation
Skin lesions in CCC typically presented on the first day of life (51 of 63 [81%]; Table 1). Occasionally onset was delayed for a few days, as late as day 6.22,28 The typical rash consisted initially of a generalized eruption of 2- to 4-mm erythematous macules, papules, and/or pustules, often on a 5- to 10-mm erythematous base. Generalized or patchy erythema sometimes was present as the initial finding. Over a few days, the typical eruption evolved to include pustules, vesicles, and, sometimes, bullae. Lesions of various morphologies, representing different stages of evolution, often coexisted. The back, extensor surfaces of the extremities, and skin folds had the greatest degree of involvement. The diaper area was relatively spared. Pustules usually were present on the palms and soles, but the oral mucosae only rarely had thrush.21 Onychia and paronychia occurred occasionally; rarely, CCC was limited to the nails.36 One 37-week-gestation neonate weighing 2395 g developed a burn-like dermatitis on the extremities like that typically seen in extremely premature infants (see below); systemic infection followed.37 Skin lesions generally resolved with desquamation within 1 to 2 weeks; lesions on the palms and soles often persisted the longest. Affected nails normalized as the nail plate grew out, and paronychia also self-resolved.
Premature infants, particularly those of approximately 24 to 26 weeks gestational age (weighing <1000 g) had a variable presentation that included: 1) a widespread papular, pustular eruption that sometimes progressed rapidly to vesicles and bullae (4 of 15 [27%])5; 2) diffuse erythematous macular patches, resembling a burn, which tended to desquamate and/or become eroded (9 of 15 [60%])3,22; 3) multiple denuded areas at birth (1 of 15 [7%])13; or, 4) a diaper dermatitis that resembled that typically associated with postnatally acquired (ie, neonatal) Candida (1 of 15 [7%]).3 Categories 2 and 3 likely represent a spectrum of a single, and the most common, presentation in this age group.
Systemically infected infants (ie, positive blood, urine and/or cerebrospinal fluid cultures) most commonly presented with respiratory distress and elevation of the white blood cell count (Table 1) with a left shift, reaching the level of a leukemoid reaction (eg, 120 000), particularly within the first 3 days of life.35,37,47 Persistent hyperglycemia and glycosuria also were reported.37 Chorioamnionitis and/or funisitis were reported more commonly in neonates who weighed <1000 g (5 of 15 [33%]) than in those who weighed more (10 of 48 [21%]).
Outcome
Neonates born with CCC and weighing <1000 g were at greater risk for development of systemic infection with Candida spp (10 of 15 [67%]) and death (6 of 15 [40%]) compared with those born weighing >1000 g (5 of 48 [10%]; 4 of 48 [8%], respectively). Systemic infection was present in 8 of 11 (73%) extremely low birth weight neonates3,13,22 and 1 of 1 term infants with burn-like dermatitis and/or multiple eroded areas. Among neonates weighing <1000 g at birth who died, all had burn-like dermatitis.3,22,13 Overall, the death rate in neonates with burn-like dermatitis (6 of 11 [55%]) was higher than in patients with the typical papulo-pustular presentation (4 of 52 [8%]).
| |
DISCUSSION |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
Risk Factors
CCC occurs in the setting of candidal vulvovaginitis, which is present in 20% to 25% of all pregnant women.48,49 Despite its frequent presence in the vagina, Candida can be demonstrated on the fetal surface of <1% of placentas.50 Extension of Candida to cause cutaneous and/or systemic congenital infection of the fetus is rare (Table 1). The reason that infants born to some women with vulvovaginitis develop CCC, while others do not, is unknown. A principal risk factor for early preterm birth with CCC appears to be the presence of a foreign body, specifically, an IUD or cervical sutures.
Diagnostic amniocentesis has preceded, and may trigger the development of CCC.51 Maternal age, parity, diabetes, urinary tract infection, or prolonged labor; prolonged rupture of fetal membranes; tocolytic therapy; or antibiotic or corticosteroid administration do not appear to be risk factors for CCC.47 On the other hand, congenital candidiasis may precipitate preterm labor and premature rupture of membranes.
Differential Diagnosis
The differential diagnosis of CCC is extensive. Neonatal candidiasis is acquired during passage through an infected birth canal, rather than in utero as in CCC. Lesions of neonatal candidiasis appear after the first week of life, and most commonly include thrush, perianal, and diaper dermatitis. A subset of neonatal candidiasis, recently coined invasive fungal dermatitis, presents in extremely premature infants with erosive, crusted plaques.52 Extremely premature infants with CCC most commonly have a similar presentation, however, ranging from a diffuse burn-like dermatitis to desquamation and erosions.3,13,22
A number of additional entities may appear during the first week of life and resemble CCC, including impetigo, staphylococcal pustulosis, Listeria monocytogenes infection, herpes simplex virus infection, varicella zoster virus infection, enteroviral infection, syphilis, erythema toxicum neonatorum, transient neonatal pustular melanosis, milia, miliaria, congenital ichthyosiform erythroderma, Ritter's disease, Leiner's disease, incontinentia pigmenti, epidermolysis bullosa, and Langerhans cell histiocytosis, among others. When evaluating umbilical cord lesions, the principal alternative disorder to consider is L monocytogenes infection, which also can produce white plaques on the surface of the cord.
Diagnosis
Definitive diagnosis of CCC is made by microscopic demonstration of spores and pseudohyphae of C albicans in skin scrapings, and culture of the organism from lesions, including macular erythematous patches and erosions in extremely premature infants. Because the organism generally is confined to epithelial surfaces (ie, skin, gastrointestinal tract), cultures of blood, urine and cerebrospinal fluid usually are negative. Cultures should be obtained from these sites, however, whenever systemic disease is suspected, and in all preterm infants. Immunologic evaluation generally fails to reveal any specific defect, regardless of whether systemic infection develops, and is of no diagnostic utility.
Definitive diagnosis of CCC also can be reached by demonstrating chorioamnionitis or funisitis. Criteria for amniotic infection with Candida spp include exclusive presence of Candida in lesions, characteristic lesions of fetal adnexae, and clinical manifestations of candidiasis in the neonate at birth or within the first week of life. Involvement of the umbilical cord is the most common presentation of candidal placentitis.29,47 Gross pathology of the umbilical cord is characteristic, consisting of discrete, round, yellow, 0.5- to 2-mm lesions in clusters along the cord, adjacent to the funicular vessels. The lesions may be macular or may project above the surface of the cord, but typically are not palpable; some penetrate deeply into Wharton's jelly. The fetal surface of the extraplacental membranes frequently is covered with a diffuse yellow exudate; sometimes yellow lesions are clustered on the amnion around the insertion site of the umbilical cord. Rarely, the placental chorionic surface, intervillous space, and chorionic villi are involved. Microscopic examination of the placenta or umbilical cord may be necessary to demonstrate infection not evident on gross inspection. Organisms in biopsy specimens from umbilical cord lesions are best visualized using periodic acid-Schiff or Gomori methenamine silver stains. The inflammatory infiltrate is wedge-shaped, with the broad base at the surface of the cord, and is composed of a mixture of neutrophils, lymphocytes, and histiocytes.29 The greater, and the deeper one looks in, the accompanying inflammatory infiltrate, the fewer the organisms present. The infiltrate does not extend to the vascular wall, but fungal elements have been demonstrated once within the umbilical vein.53 The overlying amniotic membrane is intact in lesions of recent onset, but may become necrotic as the lesion evolves and the inflammatory infiltrate intensifies.
Pathogenesis
CCC appears to be acquired in utero by the ascension of organisms from an infected vagina into the uterine cavity. The mechanism whereby this may occur, however, is largely unknown. This theory is founded on the observation that chorioamniotic plaques of Candida are localized to the inferior pole of the amniotic sac, closest to the vagina, underlying areas of vaginal inflammation, and sometimes opposite the internal orifice of the cervix.29 C albicans is capable of infecting and penetrating human fetal membranes or chick chorioallantoic membranes in vitro.54,55 Furthermore, mycelial filaments have been found inside Wharton's jelly of the umbilical cord without any detectable disturbance of the overlying amniotic epithelium.29 Although infection appears to occur in most cases through clinically intact chorioamniotic membranes, it has been hypothesized that congenital infection may be facilitated by subclinical rupture of membranes before delivery. Once the membranes have been penetrated, the pattern of fetal pathology suggests that organisms spread from the amniotic fluid to the skin, and into the pulmonary and gastrointestinal tract after the infected fluid is aspirated or swallowed.56 Infection is contained within the amniotic space, however, as there is no evidence that Candida is capable of passing the placental barrier to cause fungemia in the fetus57; conversely, hematogenous spread from the fetus to the placenta has been described.5
Because of a delicate balance between host defenses and fungal virulence factors, Candida typically is a commensal organism in the immunocompetent host. Candida infection of the skin generally develops in association with compromise of local and/or systemic immune defense mechanisms. One of the most important defenses against infection is an intact mucocutaneous barrier, which provides physical and biochemical defense against microorganisms. The immature, compromised epidermal barrier of the preterm infant may predispose to cutaneous infection and enable the organisms to penetrate into deeper layers of the skin to enter the bloodstream and cause invasive disease.3,13,22,37,52,58 Gross compromise of the epidermal barrier, as in those with burn-like dermatitis, is highly associated with systemic invasion.3,13,22,37 Acid proteinase secreted by Candida has keratinolytic activity and is thought to facilitate initiation of cutaneous candidiasis and assist in invasion.59,60
The propensity to develop invasive candidal disease, particularly in
preterm infants, also is attributable to their immature systemic
defenses. Although the ability of neonatal neutrophils to ingest and
kill Candida is similar to that of adults, their recruitment
to sites of infection is delayed and diminished.61 Neonatal T-lymphocytes (both
CD4+ and
CD8+) are deficient in their
ability to produce interferon 
(INF-), an important macrophage
activating factor; and neonatal macrophages are not fully activated by
INF-.62 Neonates also have diminished concentrations of
opsonins, including complement.
Outcome
In full-term infants, CCC almost uniformly follows a self-limited, benign course despite the presence of candidal chorioamnionitis and funisitis. Perhaps the infection occurs too soon relative to the time of delivery for systemic disease to develop, or the epithelial barriers and/or systemic immune defenses of term infants are sufficiently mature to prevent systemic infection from occurring. Consequently, an extensive evaluation is not warranted in these infants, although a high index of suspicion for disseminated disease should be maintained, because development of systemic involvement in previously well-appearing full-term infants with CCC has occurred.2
Preterm infants <27 weeks of gestational age and weighing <1000 g at birth are at the greatest risk for development of systemic disease in association with CCC. They may be stillborn or present early in neonatal life with in utero-acquired infection. Those with burn-like dermatitis are at particularly high risk for systemic infection and death.3,13,22,37
Treatment
Given the benign nature of CCC in full-term infants, there has been no proven benefit to interventional therapy. Based on anecdotal experience alone, however, topical and oral therapy has been recommended by some authors to decrease the number of viable organisms on the skin and in the gastrointestinal tract and, presumably, to lower the risk of systemic spread.18 Our patient received no antimicrobial therapy and did well, despite the widespread nature of her eruption. Perhaps future studies can examine the beneficial effects of topical and/or oral therapy in term infants with CCC.
Systemic antifungal therapy should be considered for all infants with CCC who have respiratory distress in the immediate neonatal period and/or laboratory signs of sepsis such as an elevated leukocyte count with an increase in immature forms or persistent hyperglycemia and glycosuria. Therapy should be given to neonates with CCC manifest as burn-like dermatitis, and those with positive blood, urine and/or cerebrospinal fluid cultures for Candida spp. Extensive instrumentation in the delivery room and invasive procedures such as placement of an indwelling catheter in the neonatal period; or an altered immune response, particularly neutrophil or macrophage function; may increase the risk for subsequent development of systemic disease. Prompt initiation of antifungal therapy appears to be the most important factor associated with survival of systemic infection.3,13,37 This could not be ascertained by our review, however, due to lack of sufficient detail in the majority of reports. Amphotericin B is the first-line agent for treatment of systemic disease.63 Some adjunctive benefit may be provided by 5-flucytosine. Use of fluconazole may be considered when toxicity to amphotericin B is prohibitive and the organism is susceptible, although data to support its use for treatment of systemic candidal infection in preterm infants are lacking.
| |
CONCLUSIONS |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
CCC is a rare disorder that results from a Candida spp infection acquired in utero. A major risk factor for development of congenital Candida infection is the presence of an IUD. Although all infants are at risk for disseminated disease, particular vigilance is warranted in preterm infants who are predisposed to development of invasive disease because of their immature, compromised mucocutaneous barrier and systemic host defenses. Reporting bias must be considered when attempting to draw conclusions based on an analysis of published accounts. Nevertheless, our review suggests that those at greatest risk for a poor outcome attributable to CCC are neonates <27 weeks gestational age and weighing <1000 g at birth. Those with burn-like dermatitis should be considered systemically infected until proven otherwise. Prompt recognition of CCC and early administration of systemic antifungal therapy are paramount in the management of these infants. The majority of neonates with CCC, particularly those who are full-term, follow a benign course, require no therapy, and recover with no long-term sequelae.
, §,
* Division of Infectious Disease and Division of Dermatology
Department of Pediatrics
Children's Hospital and Regional Medical Center
§ Division of Dermatology
Department of Medicine
University of Washington School of Medicine
Seattle, WA 98105
Department of Pediatrics
Group Health Cooperative
Seattle, WA
| |
FOOTNOTES |
|---|
Received for publication Feb 18, 1999; accepted Jul 20, 1999.
Address correspondence to Gary L. Darmstadt, MD, Division of Infectious Disease, Department of Pediatrics CH-32, Children's Hospital and Regional Medical Center, 4800 Sand Point Way NE, Seattle, WA 98105. E-mail: gdarms{at}chmc.org
| |
ABBREVIATIONS |
|---|
CCC, congenital cutaneous candidiasis; IUD, intrauterine device.
| |
REFERENCES |
|---|
|
Top
Abstract
Introduction
Methods
Results
Discussion
Conclusion
References
|
|---|
report of two cases.
Chang Gung Med J.
1994;
17:63-67 
1000-gram neonate.
Pediatrics
1995;
95:682-687 activation of macrophages in newborn infants.
J Immun.
1994;
153:5643-5649 [Abstract]This article has been cited by other articles:
 |
|
 |
|
  M. M. Avram, V. Gobel, and A. Sepehr Case 30-2007 -- A Newborn Girl with Skin Lesions N. Engl. J. Med., September 27, 2007; 357(13): 1327 - 1335. [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kaufman and K. D. Fairchild Clinical Microbiology of Bacterial and Fungal Sepsis in Very-Low-Birth-Weight Infants Clin. Microbiol. Rev., July 1, 2004; 17(3): 638 - 680. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Melville, S. T. Kempley;, G. L. Darmstadt, J. G. Dinulos, and Z. Miller Treatment of Invasive Candida Infection in Neonates With Congenital Cutaneous Candidiasis Pediatrics, July 1, 2001; 108(1): 216 - 216. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2000 American Academy of Pediatrics. All rights reserved. |
||
|
||
