Perinatal and childhood morbidity and mortality in congenital analbuminemia

Abstract

Albumin, a serum transport protein, provides 80% of colloid osmotic pressure. Congenital analbuminemia (CAA) is an autosomal recessive disorder characterized by absence of serum albumin. Fifty cases of CAA have been reported throughout the world; however, little is known about its clinical impact. Most reported cases have few clinical signs and symptoms. Twelve local cases from the northwestern central plains region in Saskatchewan were identified and reviewed to ascertain morbidity and mortality related with CAA. All the cases are from two remote First Nations communities. Cases had frequent hospital admissions and recurrent respiratory tract infections. Placental abnormalities included hydropic placentas, placental infarcts and microcalcifications. One-half of the cases were born preterm and one-quarter were small for their gestational age. There were three mortalities in the case series. The present case series suggests increased morbidity and mortality during infancy in patients with CAA. The long-term risks of CAA in this population are unknown and a longitudinal study is recommended.

Congenital analbuminemia (CAA) is an autosomal recessive disorder characterized by very low, or absent, serum albumin in the absence of hepatic dysfunction, renal losses or gastrointestinal losses. Albumin maintains colloid osmotic pressure and is an important carrier protein of nutrients, wastes and hormones (1). Previously reported CAA cases describe individuals as being relatively asymptomatic, with the most frequently associated complications including lipodystrophy, hypercholesterolemia, frequent lower respiratory tract infections (LRTI) in children and low birth weight (1). Few reports have provided antenatal and maternal history, although it has been hypothesized that the low incidence of CAA may in part be due to affected fetuses creating an inhospitable in-utero environment (1,2).

CAA is a rare condition with 50 cases having been recorded since 1954, with an estimated frequency of 1 in 1,000,000 (3). Cases have been widely distributed throughout the world (3). Four cases of CAA from the northwestern central plains region in Saskatchewan have been previously reported (4–6). Two of these cases underwent molecular diagnostic testing of the albumin gene and were positive for the Kayseri defect (AT-nucleotide deletion in exon 3 leading to a premature stop codon and a nonfunctional truncated protein) (5). We have identified eight additional cases from the same region. All 12 cases are of Cree descent. Anthropological studies of this region report that between 1870 and 1960 the population was primarily sustained by within group marriages, suggesting that the elevated incidence of CAA in the region is due to a founder effect (6,7). The following case series will report on the morbidity and mortality associated with CAA.

METHODS

The present study was approved by the University of Saskatchewan (Saskatoon, Saskatchewan) Biomedical Research Ethics Board. Cases were identified by searching hospital health records for admissions coded for hypoalbuminemia (E88.0) between 2001 and 2009. These charts were reviewed to identify patients with CAA (defined as serum albumin levels <10 g/L or when measured with protein electrophoresis albumin levels approximately 0 g/L in the absence of secondary causes). Medical genetics files of cases and maternal health records were reviewed for additional medical history wherever possible. Twelve cases were identified and a total of 11 cases are included in the case series. Adequate records for the one excluded case could not be obtained.

Data collection included antenatal/obstetrical and birth histories, placental pathology, frequency of lower respiratory tract infections (LRTI), hospitalizations, comorbid conditions, laboratory investigations, clinical signs and symptoms suggestive of analbuminemia and cause of death.

RESULTS

All cases were born between 1996 and 2009, and the majority of the cases were diagnosed in infancy, with many having edema or signs of fluid retention (Table 1). Protein electrophoresis showed a typical pattern of reduced albumin levels and an increase in other protein fractions (data not shown). Serum cholesterol levels were not available.

TABLE 1.

Case characteristics

Case	Age at diagnosis	Serum albumin (g/L)*	Clinical findings at diagnosis
1	2 months	3	Edema in neonatal period
2	2 months	3	Mild peripheral edema and excessive fluid retention/weight gain
3	12 months	2	None
4	5 months	6	None
5	2 days	<10	Edema and excessive fluid retention/weight gain
6	1 day	4	Edema and fluid retention/weight gain
7	3 days	4	Scrotal edema
8	2 days	8	Seizures secondary to hypocalcemia and hypomagnesemia
9	15 months	11	Pleural effusions and ascites
10	2 days	<10	Mild generalized edema, progressing to significant edema and respiratory distress requiring ventilation
11	2 days	<10	Hypotonia and fluid retention/weight gain

^*Serum albumin levels were measured by protein electrophoresis with the exception of cases 5,10 and 11 (serum analysis)

A review of the antenatal histories revealed seven mothers (three mothers had more than one affected child) with a total of 40 pregnancies and seven miscarriages. On average, the rate of miscarriage was 17.5%, consistent with population risk estimates of 10% to 25% (8). There was one near-term stillbirth, where the pregnancy was additionally complicated by poorly controlled gestational diabetes. The most commonly reported abnormality on fetal ultrasound was oligohydramnios occurring in 27% (3 of 11).

The preterm birth rate (<37 weeks’ gestational age) among cases was 63.6% (7 of 11) compared with the 2007 Saskatchewan preterm birth rate of 7.5% (9). In addition, 63.6% (7 of 11) were low birth weight infants (<2500 g) and 27.3% (3 of 11) of the infants were small for their gestational age (SGA) (ie, weight less than 10th percentile for gestational age).

Placental pathology was available in five cases. All the placentas were reported to be edematous and to weigh an average of 40% of the infant’s birth weight compared with the usual 20% (10). Microcalcifications or infarctions were also identified (Table 2).

TABLE 2.

Antenatal history

Case	BW (percentile)	Gestational age (weeks)	Ultrasound findings	Placental pathology	Pregnancy comorbidities
1	3	40	Oligohy	N/S	None
2	<3	37	No concerns	N/S	Incomplete record
3	25	34	No concerns	N/S	Placenta previa
4	20	34	No concerns	N/S	GHTN
5	15	32	Thick placenta	Large	None
6	5	35	No concerns	Large	None
7	20	39	No concerns	N/S*	Unknown
8	15	38	Oligohy	Large, microcalcif	GDM
9	75	40	No concerns	N/S	None
10	50	32	Poor BPP	Large; infarct	Type 2 DM
11	50	32	Oligohy	Large; infarct	GDM, nephro synd

^*Placenta required manual extraction under general anaesthetic. BW Birth weight; BPP Biophysical profile; DM Diabetes mellitus; GDM Gestational diabetes mellitus; GHTN Gestational hypertension; nephro synd Nephrotic syndrome; microcalcif Microcalcifications; N/S Not sent; Oligohy Oligohydramnios

Six of the cases had recurrent admissions for LRTI. Admissions were primarily clustered in infancy and some required paediatric intensive care (Table 3). There were three mortalities in the case series that were secondary to overwhelming infections, of which two were percipitated by a LRTI.

TABLE 3.

Medical history

Case	NICU Admission(s)	Recurrent LRTI*	PICU Admission(s)	Other medical diagnosis
1	Yes	Yes	No	Symptomatic neonatal hypoglycemia, chronic otitis media, asthma
2	No	No	No	Milk protein allergy
3	Yes	No	Yes	Epilepsy and developmental delay, chronic otitis media, 2nd degree burns
4	Yes	Yes	Yes	Persistent patent ductus arteriosus, bilateral grade III intraventricular hemorrhage, death
5	Yes	Yes	Yes	Neurometabolic disorder, bilateral cataracts, sensorineural hearing loss, cerebral palsy, death
6	Yes	Yes	Yes	Neurometabolic disorder, symptomatic vitamin D deficiency, death
7	Yes	Yes	No	Developmental delay
8	Yes	No	Yes	Hemangioendothelioma
9	No	No	No	Persistent patent ductus arteriosus
10	Yes	Yes	Yes	Pulmonary hemorrhage and chronic lung disease, gross motor developmental delay
11	Yes	No†	No	None

^*Recurrent lower respiratory tract infection (LRTI) is defined as ≥3 LRTI requiring hospitalization before 1 year of age;

^†This case was less than 1 year old at the time of the case review. NICU Neonatal intensive care unit; PICU Pediatric intensive care unit

DISCUSSION

Individuals with CAA undergo a series of compensatory mechanisms including an increase in alternative proteins, an increase in protein half-life and adjustments to the osmotic and hydrostatic pressure gradients (1). In the absence of serum albumin, hepatic cells compensate by increasing the biosynthesis of other plasma proteins, including α, β and γ proteins (5). These compensatory proteins raise serum total protein levels and contribute to osmotic pressure and also assume some of the roles of albumin. For example, bilirubin will alternatively bind to high-density lipoproteins in the absence of albumin (11). If serum albumin levels are restored to their normal range, by albumin infusion, this compensatory up-regulation stops and compensatory protein levels return to normal. Studies have found that the half-life of infused albumin in individuals with CAA is prolonged by up to four-times, with the rate of degradation slowing as serum albumin levels decrease (2,12,13). The osmotic pressure gradient is maintained by the compensatory increase in plasma proteins, as well as by increased transcapillary fluid filtration and lymph drainage, and the reduced hydrostatic pressure gradient is achieved by lowered capillary blood pressure and increased interstitial hydrostatic pressures (1).

Existing literature provides little information on the effect of fetal CAA in pregnancy, with reports including SGA, placental edema or infarction, and frequent miscarriage (1,13–16). The present case series supports reports of SGA infants. Epidemiological studies have found that the mean birth weight of First Nations infants is higher compared with other ethnicities and, therefore, we may have underestimated the proportion of cases born SGA, because growth percentiles were determined using Fenton growth charts (17). The rate of miscarriage was not elevated; however, there may be an underestimation of the miscarriage rate because the accuracy of the numbers depend on self-reporting, and first trimester miscarriages are often unrecognized. The finding that there were a large proportion of preterm births among these cases was a novel finding for CAA.

None of the mothers in the case series are known to have CAA. We hypothesize that when the CAA adapted fetal circulation comes into contact with the nonCAA affected maternal circulation, the hydrostatic and oncotic pressure differences may affect placental function and/or integrity, and create a suboptimal in-utero environment. The large edematous placentas and frequent SGA infants in this case series support this hypothesis.

LRTI are a significant cause of morbidity among children (18). Recurrent LRTI, particularly in infancy, were frequent within our series. Recent studies report that American Indians and Alaskan natives less than one year of age continue to have 3.6 times as many LRTI as non-native infants (19). Similar results are reported in Canadian studies (20). The frequency of LRTI in our local population was not available for comparison. Comorbid conditions including neurological disorders and chronic lung disease were also common and are risk factors for LRTI in infancy. We cannot conclude that the high frequency of LRTI and associated morbidity is entirely due to CAA; however, an association between CAA and increased LRTI is suggested.

Cases frequently had signs of fluid retention at diagnosis of CAA; however, these clinical signs often emerged after admission to hospital. We speculate that compensatory changes in hydrostatic pressure may be easily overwhelmed with the administration of isotonic fluids. Serum calcium and magnesium are carried by albumin, and there is one reported case of hypomagnesemia and tetany in the literature (1). In the present series, three cases had documented hypocalcemia and/or hypomagnesemia, of which, two cases were asymptomatic and one had seizures. Also, one case was diagnosed as being vitamin D deficient

The number of cases of CAA in this region of Saskatchewan has increased over the last decade. This increase may have occurred for a number of reasons. Historical marriage patterns by the residents of this region have likely increased the carrier frequency. Furthermore, Saskatchewan Aboriginal women have a higher fertility rate compared with nonAboriginal women (2.6 births per woman and 1.8 births per woman, respectively) (21). Because hospital records were only searched over the last decade and the majority of admissions occured in infancy, some older cases may have been missed. Increased awareness of CAA locally may also contribute to the increased frequency.

All cases in the series were identified during admission to a tertiary care hospital; therefore, it is possible that there may be asymptomatic and undiagnosed cases of CAA within the community, causing us to overestimate the associated morbidity in this population. A second limitation of our case series was the retrospective nature of the study, which meant that health records were incomplete at times. Finally, several of the cases had other medical conditions, including neurometabolic disease, epilepsy, patent ductus arteriosus, and chronic lung disease, which likely increased their morbidity and mortality.

CONCLUSION

The present case series suggests that among hospitalized patients there is a high morbidity and mortality associated with CAA. High rates of preterm birth, SGA and LRTI suggest that albumin plays an important role in placental function, development of the respiratory system and immunity. All cases received albumin infusions; however, there were insufficient data to determine if this had an effect on morbidity or mortality. The hospitalized child with CAA may have increased susceptibility to fluid overload, hypocalcemia and hypomagnesemia. We are not able to make any conclusions about management of CAA patients or their long-term health risks. We support collaboration between local communities and academic centres to identify priorities to optimize care and determine the longitudinal implications of CAA.