A secondary data analysis was conducted to estimate the cumulative incidence of infants with LCAs in six hospitals of Lima, the department of Lima and throughout Peru. For its part, the INMP carried out a retrospective analysis of the medical records of neonates born with congenital anomalies to describe their characteristics.

According to SEGO, which proposes a list of 17 of anomalies (Table 1), LCAs are defined as conditions that, due to their severity, do not require evaluation by a clinical committee to determine its classification, since it would be considered as such anywhere in the world due to its poor prognosis. ¹ However, since the records analyzed were based on the 10th revision of the International Classification of Diseases (ICD-10), ¹⁶ it was only possible to evaluate the 9 LCAs currently included in that classification. It should be noted that the term LCA has not yet been adopted by the Peruvian health system.

Table 1

N	Diagnosis	ICD-10 Code	Evaluated in this study
1	Anencephaly/ Acephaly/Acrania	Q00.0	Yes
2	Holoprosencephaly	Q04.2	Yes
3	Renal agenesis, bilateral	Q60.1	Yes
4	Potter's syndrome	Q60.6	Yes
5	Thanatophoric short stature	Q77.1	Yes
6	Trisomy 18	Q91.0 -Q91.3	Yes
7	Trisomy 13	Q91.4 -Q91.7	Yes
8	Trisomy 9	Q92.0 -Q92.1	Yes
9	Triploidy and polyploidy	Q92.7	Yes
10	Hydranencephaly	-	No
11	Laryngeal atresia	-	No
12	Tracheal atresia	-	No
13	Agenesis of the diaphragm	-	No
14	Ectopia cordis	-	No
15	Pentalogy of Cantrell	-	No
16	Amniotic band syndrome	-	No
17	Limb-body wall complex	-	No

ICD-10: International Classification of Diseases, 10^th revision.

A newborn with LCA was that which had one of the 9 LCA diagnoses contemplated in the ICD-10. Anomalies that were not coded in the ICD-10 were not included in the study as it was not possible to identify them. Cumulative incidence was used as a unit of measurement; it was estimated by dividing the number of reported LCA cases by the number of live births in the period studied, which, for accessibility reasons, varied as follows: from 2011 to 2015 in the Hospital Nacional Arzobispo Loayza (HNAL), the Hospital Nacional Docente Madre Niño San Bartolomé (HONADOMANI) and the records from Peru and Lima, and from 2012 to 2016 in the Hospital Nacional Cayetano Heredia (HNCH), the Hospital María Auxiliadora (HMA), theHospital Nacional Sergio E. Bernales (HNSEB) and the INMP.

Information on the incidence of the 9 LCAs contemplated in the ICD-10 was collected over a 5-year-period for all of Peru, the department of Lima, 5 hospitals in Lima and the INMP from three sources:

For all of Peru and for the department of Lima. This information was requested from the MINSA's Public Information Access System (SAIP),¹⁵ which obtains its data from the hospital discharge records of the ministry's level II and III centers and merges them into the central statistics office. It should be noted that only the main diagnosis of each patient's epicrisis is found in these records. In addition, the total number of births reported to MINSA for all of Peru and the department of Lima was requested in order to calculate incidences.

For hospitals in Lima. This information was requested from the SAIP for each of MINSA's Level III hospitals in Lima, and only five responded: the HNCH, the HNAL, the HMA, the HONADOMANI and the HNSEB. Only the main diagnosis of the epicrisis of each newborn is found in these records. In addition, the number of births reported by each of the hospitals assessed was requested in order to calculate incidences.

For the INMP. The INMP's Statistics and Information Office was asked for the hospitalization databases of its neonatal service, in which the ICD-10 codes of each newborn born in that hospital were recorded, including stillborn children. After identifying the neonates with any of the 9 LCAs described in the ICD-10, a manual review of their medical records was performed to corroborate the diagnoses and extract the demographic and maternal/perinatal characteristics.

The main variable considered for the present study was the presence of any of the LCAs evaluated as described above. In the specific case of INMP, other variables collected during the review of medical records were: mother's age, gestational age at birth using the Capurro test (full term ≥37 weeks of gestation), type of delivery (vaginal or cesarean), neonatal depression (score <7 on the Apgar score 5 minutes after birth), sex of the newborn, hospitalization and death during the hospital stay.

Central tendency measures were used for the presentation of the results: dispersion for quantitative variables, and relative and absolute frequencies for qualitative variables. The analysis was done using Stata v14.0 (Stata Corp, College Station, TX, US).

This study followed the ethical principles outlined in the Declaration of Helsinki.¹⁷ The research was based on secondary database analysis of, which were analyzed respecting privacy.

The protocol of the present study was approved by the INMP's Institutional Committee of Ethics through letter No. 0213-2017-DG-N°-083-OEAIDE/INMP of September 20, 2017.

The incidence of LCAs found in the records for Peru, for the department of Lima and for most of the hospitals evaluated was lower than that found in the INMP database. This may be explained by differences in the methodology used, since INMP data were obtained from the hospital database, while central and hospital reports were used for the other populations evaluated (Peru, Lima department, and the other hospitals) and they only included the main diagnosis recorded in the patient's epicrisis at discharge, which may not necessarily be a LCA.

Other explanations to this difference in the incidence would be that other medical centers refer pregnant women with fetuses diagnosed with some LCA to the INMP, thus increasing its numbers. This could also be associated with the fact that the INMP, which has the largest neonatology service of the country, has trained personnel and the necessary supplies to make an adequate diagnosis of LCA, which would be underdiagnosed in other places.

On the other hand, the incidence for Peru (0.89 LCAs per 10 000 births) was much lower than that reported by international surveillance systems such as EUROCAT,⁶ the Latin American Collaborative Study of Congenital Malformations (ECLAMC), ⁹ the National Registry of Congenital Anomalies of Argentina (RENAC) ¹⁸ and the Mexican Program for Registration and Epidemiological Surveillance of External Congenital Malformations (RYVEMCE), ¹⁹ which had incidents between 6.60 and 14.53. Since there is no reason to assume that the incidence of LCAs in Peru is lower than in other countries, estimates are that the records provided by the SAIP underestimate the figure by about 90%.

All this is very concerning, since knowing the actual magnitude of the problem, detect clusters of cases, carry out studies of associated factors and evaluate the impact of preventive measures, such as the use of folic acid supplements to avoid neural tube defects, is only possible if LCAs are properly reported.²⁰ This could be achieved by adopting an anomalies surveillance system as is the case of other countries and regions. ^6,9,18,19

Regarding international surveillance systems, the incidence reported by EUROCAT was twice as high as that reported by ECLAMC, RENAC, RYVEMCE and INMP. The reason may be that EUROCAT reports congenital anomalies in hospitals of various European countries in a standardized manner, while the other systems tend to combine hospitals and have varying degrees of standardization in their reporting. It is also possible that population differences, such as maternal age at conception and the spread of prenatal diagnosis in Europe, make EUROCAT records larger than those of other surveillance systems. Table 4 presents the incidents found by the above-mentioned international surveillance systems.

Table 4

Diagnosis	Peru (20112015)	INMP (20122016)	RENAC (20122015)	EUROCAT (20112015)	ECLAMC (20072011)	RYVEMCE (20072011)
Anencephaly/Exencephaly/ Acrania	0.71	3.59	2.75	4.04	5.79	3.58
Trisomy 18	0.13	1.80	1.13	5.67	1.32	0.67
Alobar holoprosencephaly	-	0.90	2.38	1.51	0.78	1.79
Potter's syndrome	-	0.75	-	1.22	-	-
Trisomy 13	0.06	0.15	0.40	2.09	0.57	0.56
Bilateral renal agenesis	-	-	0.43	-	-	-
Thanatophoric dysplasia	-	-	0.12	-	-	-
Trisomy 9	-	-	-	-	-	-
Triploidy	-	-	-	-	-	-
Total	0.89	7.19	7.21	14.53	8.46	6.60

INMP: Instituto Nacional Materno Perinatal; RENAC: National Registry of Congenital Anomalies of Argentina; EUROCAT: European Surveillance of Congenital Anomalies; ECLAMC: Latin American Collaborative Study of Congenital Malformations; RYVEMCE: Mexican Program for Registration and Epidemiological Surveillance of External Congenital Malformations.

When reviewing the medical records of infants with LCAs born at INMP, it was found that 4 out of 10 died before they could be hospitalized, either in the delivery room or in the newborn's immediate care area, and that a similar proportion died during hospitalization. It should be noted that this high mortality is to be expected for this type of anomalies

When the different types of LCAs are studied separately, it can be seen that survival at discharge is almost entirely attributable to cases of holoproscencephaly and trisomy 18. Previous studies have also found a high survival rate related to these pathologies in the first week (71% and 65%, respectively), which drops drastically after the first year (47% and 16%, respectively). ⁽² This relatively long survival could impact the mother and the rest of her family financially and mentally, and this should be evaluated in future studies.

Most newborns with LCAs die, and if they survive, the degree of disability is high. Although these types of anomalies are rare, they are highly relevant because of their potential impact on mothers' physical and mental health, such as psychological reactions of hopelessness, sadness, or guilt.^21-23 The magnitude of these damages has not been adequately assessed in countries where termination of pregnancy due to LCAs is illegal, and more studies are needed in Peru to assess the consequences and effectiveness of preventive interventions -such as the use of folic acid supplements for the prevention of anencephaly²⁴ or appropriate counseling on the mother's age at conception as a risk factor for trisomy 13 or trisomy 18-²⁵ and recovery interventions -including psychological support to the woman during and after pregnancy and avoiding medical futility in these babies.

The main limitation of this study was the use of a secondary database (SAIP) and clinical records for the collection of information, which prevents ensuring that all diagnoses are reported, or that all the reported diagnoses are entered into the databases of the centers from which the information was obtained. Furthermore, since these databases only included the LCAs covered by the ICD-10,¹⁶ it was not possible to track the 17 LCAs proposed by SEGO;¹ therefore, the cumulative incidence of LCAs is expected to be underestimated.

Another limitation is that the SAIP may have duplicate data, i.e. some patients may have gone to more than one hospital. However, due to the high lethality of LCAs during hospitalization and the underreporting observed, this was considered unlikely.

This is the first study that describes the incidence and characteristics of LCAs in Peru, so it is a relevant source of information to understand the impact and consequences that these anomalies can have in the country.