LCAT Deficiency in Amerindian Populations: A Systematic Review With the Clinical and Genetic Description of Mexican Kindred


 Background: LCAT deficiency is a rare disease, characterized by two distinct phenotypes, familial LCAT deficiency (FLD) and Fish Eye disease (FED). There is little knowledge of LCAT deficiency syndromes in Amerindian populations. We present the results of the first systematic review evaluating the ethnic distribution of LCAT deficiency, with particular emphasis on Latin America and discuss the case histories of three Mexican-Mestizo probands. Methods: A systematic review was conducted following the PRISMA Statement in Pubmed and SciELO. Articles which described subjects with LCAT deficiency syndromes and an assessment of the ethnic group to which the subject pertained, were considered for analysis. Results: In our region, 47 cases of LCAT deficiency have been published from six countries (Argentina (1 unclassified), Brazil (38 FLD), Chile (1 FLD), Columbia (1 FLD), Ecuador (1 FLD) and Mexico (4 FLD, 1 FED and 1 unclassified). In Mexico, one of the FLD probands’ showed a novel mutation; this patient came from an isolated village in the south of Mexico, with little genetic admixture in this region.The systematic review revealed 215 cases of LCAT deficiency (154 FLD, 41 FED and 20 unclassified) in at least 33 ethnic/racial groups (predominantly Caucasian). In addition, at least 138 different mutations in the LCAT gene have been identified. There was no association between genetic alteration and ethnicity. The mean age of diagnosis was 42 ± 16.5 years, with FED identified significantly later than FLD (55 ± 13.8 vs. 41 ± 14.7 years respectively). The prevalence of premature coronary heart disease was significantly greater in FED vs. FLD (p=0.00). Conclusion: The systematic review shows that LCAT deficiency syndromes are clinically and genetically heterogeneous. We were unable to confirm any association between ethnicity and LCAT mutation. However, we were able to show a significantly greater risk of premature coronary artery disease in FED compared to FLD. In FLD, the emphasis should be in preventing progression of renal disease, while in FED, cardiovascular risk management should be the priority. The LCAT mutations discussed in this article are the only ones reported in the Mexican- Amerindian population.


Introduction
Lecithin cholesterol acyltransferase (LCAT) is a 67 kDa protein, predominantly expressed in the liver (1). It circulates in plasma bound to high density lipoproteins (HDL) but can also be found on apolipoprotein B100 containing particles (2,3,4). It catalyzes the transfer of an unsaturated fatty acid from lecithin to free cholesterol, producing lysolecithin and cholesteryl ester. This reaction occurs on immature HDL particles in the presence of apolipoprotein A-I (apo A-I), and corresponds to the alpha activity of the LCAT enzyme. When this reaction occurs on LDL or VLDL it is referred to as the beta activity. The net result is the formation of hydrophobic cholesterol ester, which is transferred to the lipoprotein core. In the case of HDL, this allows the conversion of discoidal pre-beta 1 particles to mature spherical alpha forms. In addition, the esteri cation of cholesterol on HDL increases the concentration gradient for free cholesterol between cell membranes and HDL, thus promoting the removal of cholesterol from cells (1,2).
LCAT de ciency is a rare autosomal recessive disease (3,4). Loss of LCAT function causes decreased maturation of HDL particles and increased HDL levels of unesteri ed cholesterol and phosphatidylcholine. There is no reliable estimate of the prevalence of the disease; in individuals with low HDL cholesterol (HDL-C) ranges, the estimated prevalence of LCAT de ciency is between 2-9% (5,6,7). The disease is characterized by two distinct phenotypes, familial LCAT de ciency (FLD) and Fish Eye disease (FED). In FLD, both the alpha and beta LCAT activity is lost, leading to extremely low plasma HDL-C (below the 5th percentile for the population), premature corneal opaci cation, hemolytic anemia, proteinuria and renal failure (8). In FED, only the alpha LCAT activity is lost, Familial Lecithin cholesterol acyl transferase de ciency (FLD), LCAT enzyme de ciency (partial and total), LCAT gene mutation or polymorphism, homozygotes, compound heterozygotes, corneal opacities, corneal clouding, low high density cholesterol levels, anemia, renal failure, atherosclerosis. These terms were cross-referenced with the keywords and MeSH terms for ethnicity: minority groups, ethnic groups, African-Americans, Hispanic-Americans, American Native Continental Ancestry Group, Amerindian, Oceanic Ancestry Group, Black, Hispanic, Latino, Asian American, African American, Native American, Indian, Asian, and Paci c Islander. Articles written in English or Spanish were included. Data collection was carried out by four investigators, commenced in September 2018 and concluded in March 2020. The investigators took care to avoid double counting of cases. In addition, the reference lists of review articles and conference abstracts were also considered. Abstracts were independently assessed to identify eligible research reports. The commonest reasons for ineligibility were; insu cient information regarding either the LCAT gene mutation, clinical characteristics or no mention of ethnicity.

Methods For Biochemical And Genetic Analysis Of Lcat Probands
Biochemical measurements Fasting blood samples were obtained from all three probands. These included full blood count, chemistry, a complete lipid pro le, erythrocyte fragility studies (in FLD de ciency subjects), and 24 hour urine collection for determination of microalbuminuria and creatinine clearance. The lipid parameters were measured in the Institute's central laboratory. For total cholesterol, HDL cholesterol (HDL-C), LDL cholesterol (LDL-C), triglycerides and glucose measurements commercial enzymatic methods were used (Beckman Coulter). Apolipoprotein A1 and apolipoprotein B concentrations were measured using nephelometric methods (Beckman Coulter).
Measurement of LCAT activity α-LCAT activity was measured by the method of Chen and Albers (16). Brie y: apoAI/phosphatidilcholine/ 3 H-cholesterol complexes were incubated with plasma in a shaking water bath for 1 hour al 37 °C (esteri cation was linear during this time). The reaction was stopped, and lipids were extracted. Esteri ed and unesteri ed cholesterol were separated by thin-layer chromatography, and the radioactivity was counted. LCAT speci c activity was expressed as the nanograms of cholesterol esteri ed by 1 L of plasma in 1 hour (nmol/mL/h).
Measurement of PON-1 activity (patient 1 and kindred, patient 3) PON1 activity was measured using phenylacetate as substrate (17). Initial rates of hydrolysis were determined spectrophotometrically at 270 nm. The assay mixture included 1 mM phenylacetate and 0.9 mM CaCl 2 in 20 mM Tris-HCl, pH 8.0, and 10 µL serum (diluted 1:40). The 270 for the reaction was 1310 M − 1 cm − 1 . Arylesterase activity was expressed as the number of micromoles of phenylacetate hydrolyzed per minute per milliliter of serum. To determine the distribution of PON1 in lipoprotein fractions, 300 µL of plasma heparin was separated by size exclusion chromatography using a Bio-Prep SE1000/17 column coupled to a Bio-Rad Duo Flow system as previously described (18) with slight modi cations. Brie y, protein elution was accomplished with 2 mM CaCl 2 in 20 mM Tris-HCl, pH 8.0, at a ow rate of 1 mL/min. Fractions of 0.5 mL were collected and PON1 activity was assessed after elution using 10 µL of each fraction. The column was calibrated with VLDL, LDL and HDL isolated by ultracentrifugation from a pool of 5 plasma samples obtained from 5 normolipemic volunteers. For the calibration, cholesterol was determined in the elution fractions by enzymatic colorimetric methods commercially available.

Statistical analysis
The distribution of categorical variables is reported as frequencies and percentages. Continuous data is described as mean and standard deviation or with median and interquartile range depending on the parametric or non-parametric distribution of variables. Statistical analyses was performed using Statistical Package for Social Science (SPSS Inc, Chicago, IL, and Version 21.0) and GraphPad Prism, version 7.0.

Results Of The Systematic Review
The PRISMA algorithm is shown in Fig. 1. Our research strategy retrieved a total of 3,373 publications. After removing any duplicate documents, 2,800 abstracts were reviewed. Of these, 2,153 articles were excluded, as they did not complete inclusion criteria. In total, 87 relevant articles/abstracts were reviewed in detail for eligibility. Of these, six publications were excluded due to incomplete information. Finally, 81 studies were included for the purposes of this article (Table 1, ). The systematic analysis retrieved 215 cases, of which 71.6% (n = 154) were FLD, 19.0% (n = 41) were FED and 9.3% (n = 20) were unclassi ed ( Table 2). Most of the information was found in case reports (87.6%). The LCAT de ciency cases are from 33 countries, the majority of individuals are Caucasians and the commonest presenting feature was corneal opacity. There is a predominance of men (n = 116, 53.9%) and the mean age of individuals is 42 ± 16.5 years. The median concentration of HDL-C is 7 (4-12) mg/dl and median LCAT activity is 1.65 (0.0-7.1) nmol/ml/hr. A creatinine clearance < 60 ml/min was found in 30.2%, > 60 in 40.4% and unknown in the remaining cases. Albuminuria/ proteinuria was present in 39.1% and absent in 29.8% of cases. Anemia was reported in 53.9% and absent in 32.1%. Premature coronary artery disease was present in 7.4%, absent in 59.1% and not evaluated or unknown in the remaining cases. On comparing the individuals with FLD and FED, certain differences are apparent. The FLD cases are signi cantly younger than the FED cases (41 ± 14.7 vs. 55 ± 13.8 years, p = 0.02, respectively). There was no difference in HDL-C levels between groups. However, LCAT activity was signi cantly lower in FLD compared to FED (0.1 (0.0-2.1) nmol/ml/hr vs. 2.7 (0.8-7.0), p = 0.01). Unsurprisingly, clinical features compatible with FLD are signi cantly more common in these cases (low creatinine clearance, albuminuria/proteinuria and anemia). Premature coronary artery disease was signi cantly more prevalent in FED compared with FLD (p = 0.00).

Mutational analysis:
A total of 138 mutations in the LCAT gene were recovered (136 in exons and 2 in introns) (supplementary table 1). Mutations have principally been published in Caucasians. Genetic alterations are present on all exons of the gene; there was no association between a particular exon and phenotype. No speci c mutation was associated with an ethnic group. The number of mutations associated with FLD, FED and unclassi ed cases were 77, 38 and 23 respectively.
The ethnic distribution of the cases was reviewed with respect to location of LCAT mutation (supplementary table 1). Here exon 6 (n = 41) and exon 1 (n = 19) were the most common sites for LCAT mutations. There was a predominance of exon 6 mutations, in particular in Italians, Dutch and Japanese groups. In the Amerindian ethnic group, exon 1 appeared most common in Mexican-Mestizos whilst exon 6 predominated in Brazil and Chile.
Finally, the number of mutations per exon, adjusted for size of exon was examined (supplementary table 2). This avoids exon size bias; exon 6 is more than Characteristics of the Latin American cases: In total, 47 cases of LCAT de ciency have been published from six Latin American countries (Argentina, Brazil, Chile, Columbia, Ecuador and Mexico) ( Table   3). There are 38 FLD cases from Brazil (published in an abstract), one unclassi ed case from Argentina and 3 FLD cases from Chile, Columbia and Ecuador respectively. In Mexico, six cases (4 probands) have been encountered (4 FLD, 1 FED and 1 unclassi ed); one of which has previously been published (unclassi ed probable FED).  The proband was a 37-year old woman with bilateral corneal opacities (no de cit in visual acuity). She came from a small village in the state of Oaxaca, in south-west Mexico. She was the 6th of 10 children and her parents were apparently non-consanguineous. Only her paternal grandmother had eyes similar to hers. Of her 9 siblings, 2 brothers had corneal opacities and nephrotic syndrome. There was no history of cardiovascular disease in her family. We studied all available members of her family, including her parents and 5 of their 10 children.
She had a history of hyperlipidemia, arterial hypertension and nephrotic syndrome; a renal biopsy reported glomerulopathy characterized by mesangial proliferation, vacuolated macrophages and presence of intramembranous lipid deposits in glomerular capillaries. A recent carotid doppler ultrasound was normal with no alteration in carotid-intima thickness.

Biochemical analysis:
Laboratory results showed a normochromic, normocytic anemia (hemoglobin 9.2 g/dl (normal 13-15 g/dl)) and measurement of erythrocyte osmotic fragility con rmed the presence of brittle cells. There was evidence of renal failure with nephrotic syndrome (creatinine clearance 47 ml/min, and proteinuria of 5 g/24hrs). The lipid pro le showed low HDL-C, hypertriglyceridemia and low levels of apolipoprotein A1 (Table 4). The LCAT activity was low (LCAT activity 0.4%, speci c activity 3.7 nmol/ml/hr) and there was a reduction in paroxonase-1 activity (27.8, control = 100.77). Two brothers were affected (homozygotes) and heterozygote family members had half-normal HDL-C concentrations ( Table 4). All three affected individuals showed some paroxonase-1 activity, whereas LCAT activity was virtually absent. The proband had a 72% reduction in PON-1 activity, while her affected brothers showed a 47% and 57% reduction respectively. The 2 remaining siblings and both parents had low LCAT activity (2.4-3.8%) and higher paroxonase activity compared to the affected individuals. On separation of the lipoproteins by exclusion chromatography, the paroxonase-1 activity was essentially on HDL, with little activity on LDL.

Mutational analysis:
A novel mutation was encountered in this proband. This was a nucleotide replacement resulting in a stop codon at position 8 on exon 1 of the LCAT gene (in the leader sequence). Tryptophan (TGG) was replaced by Ambar stop (TAG). This is reported as Trp8* or Trp-17* (*indicates stop codon) in the nucleotide sequence. The parents were heterozygous for the mutation and the proband and both her affected brothers were homozygous. The family pedigree is shown in the supplementary Fig. 1.

PROBAND 2: Fish Eye Disease (FED)
The 70 year old proband from Mexico City, had bilateral corneal opacities and a history of myocardial infarction. The subject's father and mother had suffered from coronary artery disease (her father died at age 66, her mother died at age 70). The only other family members with similar eyes were her father, paternal grandmother and one male sibling who died soon after birth. None of her 8 siblings were alive. Both her children and 5 grandchildren had normal corneas and no health issues. We studied the proband, her two daughters and 4 of her grandchildren.
The proband had type 2 diabetes mellitus (no known complications), mixed hyperlipidemia, and arterial hypertension (history of atrial brillation and left ventricular hypertrophy). The coronary heart disease was characterized by occlusion of 3 coronary vessels (left coronary: trunk, circum ex and right coronary); she had been treated with two medicated stents.
Biochemical analysis: The lipid pro le showed an HDL-C level of 11 mg/dl ( Mutational analysis: Two mutations were found on exon 1: 1. On one allele, there was an insertion of cysteine (reported as c.101dupC) at codon 34. This mutation resulted in a stop codon 7 codons later. The proband was heterozygote for this mutation.
2. On the other allele, the alteration was c.110C > T. When this allele is translated, threonine is substituted by methionine (ACG-ATG Thr37Met) (missense mutation) at position 37 of the protein. The proband was heterozygote for this mutation.
The proband is compound heterozygote for both mutations. The family pedigree is shown in Supplementary Fig. 1. Daughter 1 is heterozygote for the second mutation (c.110C > T at codon 37), while daughter 2 is heterozygote for the rst mutation (c.101dupC at codon 34). Analysis of the apoA1 gene and its promotor region was also carried out in the proband, no alterations were found.
The family pedigree is shown in supplementary Fig. 1.

PROBAND 3 (Familial LCAT De ciency).
The proband was a 29-year-old woman from Monterrey, with bilateral corneal opacities resembling premature corneal arcus. She had an FLD phenotype with extremely low levels of HDL-C, anemia and kidney disease (glomerulopathy). There was no clinical or biochemical evidence of FLD or FED in her parents or sibling. There was no family history of premature cardiovascular disease. She had attended consultations with several specialists, had undergone three kidney biopsies and one bone marrow aspiration; despite this she had not been diagnosed.
Biochemical analysis.
The proband had an HDL-C level of 4 mg/dl ( Table 4). The laboratory pro le showed: glucose 101 mg/dl, creatinine 1.12 mg/dL, hemoglobin 12.9 g/dL and 24-hour urinary protein 2307 mg/day. The proband had extremely low LCAT activity (7.3 nmol/ml/hr, LCAT speci c activity in control 145.34 nmol/ml/hr) and a 60% reduction in PON-1 activity compared with controls.

Mutational analysis:
The genetic alteration was a point mutation in exon 4 of the LCAT gene, i.e., a G to A substitution on codon 140 converting Arginine to Histidine. The family pedigree is shown in supplementary Fig. 1.

Discussion
FLD and FED are rare LCAT de ciency syndromes with differing clinical manifestations. In this article, we discuss the results of the rst systematic analysis evaluating the ethnic distribution of LCAT de ciency, with particular emphasis on Latin America, and we present the case histories of three Mexican-Mestizo probands.
The systematic review retrieved 215 published cases of which 71.6% were reported as FLD, 19% as FED and 9.3% were unclassi ed. This number is signi cantly greater than that reported in the current literature (101). The majority of probands have been published in case reports, often with incomplete clinical or genetic information. It is evident that this disease continues to be encountered at a late age, with corneal opaci cations being the principal reason for consultation. Furthermore, FED is diagnosed signi cantly later than FLD, probably due to the more severe clinical phenotype of the later warranting earlier medical attention. Timely diagnosis of this disease is needed for the application of preventive strategies and the use of newer therapies.
The biochemical features of the cases showed that LCAT enzyme activity was signi cantly lower in FLD compared to FED. Low HDL-C levels are a characteristic feature of this disease, however, there was no difference in concentrations between the phenotypes. Indeed, FLD and FED can have similar lipid pro les, suggesting any variability in parameters is unrelated to LCAT function. Pavanello et al. have commented that the severity of the hypoalphalipoproteinemia varies widely among carriers of different LCAT genotypes (101). Furthermore, carriers of one mutant LCAT allele show an intermediate biochemical phenotype between homozygous carriers and controls, suggesting that the disease, which is reported as recessive, is indeed codominant for the biochemical phenotype. This was also con rmed in the relatives of the Mexican-Mestizo probands.
The clinical features of the cases showed a clear difference between FED and FLD, with renal disease and anemia prevalent in the later. However, not all the FLD cases showed signi cant proteinuria or a reduction in eGFR; this suggests that the rate of progression of renal failure may well be variable. Lamiquiz-Moneo et al. state that this clinical variability is likely to be related to the biochemical phenotype rather than to the inherited mutation (80). In addition, 9.3% of cases had an unclassi ed clinical phenotype; the authors could not con rm either familial LCAT de ciency (FLD) or sh eye disease (FED). Some authors have commented that the clinical manifestations of patients with LCAT gene mutations may vary even among members of the same family carrying identical mutations (41). Therefore, it is evident that LCAT de ciency syndromes show both biochemical and clinical heterogeneity.
An important nding of this systematic review was the signi cantly greater prevalence of premature CHD in FED patients compared to FLD patients. The cardiovascular risk associated with LCAT de ciency syndromes has been a matter of debate for a number of years. A severe de ciency of HDL-C in LCAT de cient carriers would be expected to increase their risk of developing coronary heart disease (83). Oldoni et al., have compared carotid intima media thickness between 33 heterozygous FLD subjects and 41 heterozygous FED subjects (102). Carriers of FLD mutations exhibited less carotid atherosclerosis, whereas those with FED mutations presented with more subclinical atherosclerosis. The authors proposed that this discrepancy was related to the capacity of LCAT to esterify cholesterol on apolipoprotein B-containing lipoproteins-this capacity is lost in FLD, but is unaffected in FED. In a study of Italian FLD families, the inheritance of a mutated LCAT genotype had a remarkable gene-dose dependent effect in reducing carotid IMT, whereas a subgroup of these carriers also showed normal ow-mediated dilation (65,83,103). A Mendelian randomization study in 54,500 subjects concluded that common genetic variation in LCAT resulting in decreased HDL-C levels, did not associate with an increased risk of ischemic cardiovascular disease (104). Low HDL-C levels robustly associated with increased risk of myocardial infarction (MI), but genetically decreased HDL cholesterol did not. This may suggest that isolated low HDL cholesterol levels do not cause MI; the inverse relation between HDL-C and CHD observed in epidemiological studies may not be causal.
The molecular defects associated with LCAT de ciency syndromes show heterogeneity. In total, 138 LCAT mutations were encountered with no particular exon dominating in a particular ethnicity. Again this number is greater than that reported in the literature (101). There was no association between clinical phenotype and genetic alteration, this may be due to the low number of cases worldwide. Exon 6 was the predominant site for both FLD and FED; however, after adjusting for exon size, exon 1 and 4 showed the greatest concentration of mutations. At present, it is impossible to predict the phenotype (FLD or FED) associated with the LCAT mutations (101).
With regards to ethnicity, at least 33 different groups are represented, of which Caucasians are most common. The predominance of Caucasian and Asian cases may re ect better health awareness and access to health care compared with developing regions. Remarkably, only one case has been found in the African sub-continent. Cases are more likely to be found in a uent countries, but also in countries hosting research groups with interest and resources to investigate this disease. In Latin American few cases have been published (82)(83)(84)(85)(86)(87)(88). On interesting nding was that the proband continued to show paroxonase-1 activity. This was essentially on HDL, even though the number of particles was extremely low and despite a clear lack of LCAT activity. The proband had 27% activity, while her affected siblings had approximately 50% activity. The heterozygote family members had essentially normal paroxonase-1 (PON-1) activity. This enzyme prevents the conversion of LDL cholesterol into a more atherogenic particle (105). Preserved PON-1 activity has been reported in other HDL de ciency states, and in vitro experiments with LCAT de cient plasma suggest an apparent maintenance of cholesterol e ux (94,106,107). Although HDL cholesterol levels were reduced by 93%, there was only a 50% reduction in reverse cholesterol transport (RCT). This suggests that RCT is conserved even in the presence of complete LCAT de ciency, supporting the differential cardiovascular risk between phenotypes.
The second Mexican-Mestizo proband (FED) had two distinct LCAT mutations, one on each allele (compound heterozygote). One of the alterations was a frameshift mutation (c.101dupC) on exon 1; the other mutation was a missense mutation (c.110C > T) on the same exon. Both mutations have previously been reported in the literature (25,26,27). Predicting the effect of the co-existence of these mutations (one on each allele) on LCAT function and structure is not straightforward. The majority of mutations are not located in sites involved in the catalytic function of the enzyme; the affected sites are probably involved in maintaining protein stability and structure. The mature LCAT protein contains 416 amino acids and a leader sequence (67 kDa) (108,109). The enzyme is thought to undergo posttranslational glycosylation which appears to be essential for the conformational stability of the protein (110). In addition, LCAT has two disul de bridges between Cyst50-Cys74 and Cys313-Cys356; the rst bridge partially covers the active site of LCAT, forms part of the lid region and is thought to enable the enzyme to bind to lipid surfaces. Hence, in this patient, the genetic alterations may interfere with the nearby lid structure or produce a conformational change when the mature protein is folded, resulting in enzyme-substrate interference. The frameshift mutation is a more detrimental alteration; however, clinical expression of which would only be apparent in homozygotes. Hence, the predominant phenotype in our subject is FED.
The third Mexican-Mestizo proband (FLD), had a point mutation on exon 4 of the LCAT gene, i.e., a G to A substitution on codon 140 converting Arginine to Histidine. This mutation has been reported previously in an Austrian kindred who were also homozygous for this modi cation (37). It appears that this domain (where Arg140 resides) is crucial for an enzymatically active LCAT protein, mutations in this region possibly affect tertiary structure.
In Latin America, persons with LCAT de ciency syndromes face unique challenges. The medical community is unaware of this condition; our third proband had attended consultations with several specialists, had undergone three kidney biopsies and one bone marrow aspiration; despite this, she had not been diagnosed. Additionally, many centers do not have the infrastructure to carry out the biochemical or genetic studies necessary to con rm this condition.
Current management of FLD is preventative and involves lipid lowering therapy, ACE inhibitors, diuretics and steroids, in order to delay progression to endstage renal disease: for many in Latin America, these medications will be an out of pocket expense. Furthermore, in this region, access to further treatment with peritoneal dialysis or hemodialysis is variable. This is related to fragmented health care coverage and socioeconomic inequality. Although, provision of renal replacement therapy (RRT) has increased in all Latin American countries over the past 20 years, universal access is available in only a few countries (Argentina, Brazil, Chile, Cuba, Uruguay, Venezuela, and Colombia) (111). Kidney transplantation may offer a temporary cure, but reoccurrence of nephropathy is inevitable and occurs within a few years (112). Currently, trials are underway with human recombinant LCAT enzyme and there is the possibility of gene therapy in the future (113). However, such products maybe subsequently unavailable and/or unaffordable (cost is much greater than average monthly income) to most of the Latin American population (114).
Earlier identi cation and adequate follow up of patients is urgently needed; the implementation of models of care and national disease registries can aid in this process. Recently, the Norwegian National Advisory Unit on Rare Disorders has been asked to establish a worldwide contact registry on FLD; this will allow the integration of efforts throughout the world to tackle the health burden and improve care for this condition (83).
We must acknowledge the strengths and limitations of this work. This is the rst systematic review of LCAT de ciency syndromes; this work highlights the major knowledge gaps in this disease. The overall number of cases and mutations is far greater than currently thought. In Latin America, the limited number of cases may have in uenced our ndings and further work is necessary in order to con rm whether isolated geographical regions may have ethnicity speci c mutations. Measurements of free cholesterol and cholesteryl ester, as well as cholesterol esteri cation rate to complete the biochemical characterization of the Mexican probands and their families would have been desirable, this was limited by the death of two of the probands, and the geographical location of the last proband.
In conclusion, the systematic review shows that LCAT de ciency syndromes are diagnosed late; with FLD cases identi ed signi cantly earlier than FED.
Furthermore, we con rm that this condition is clinically and genetically heterogeneous. We were unable to con rm any association between ethnicity and LCAT mutation. However, we were able to show a signi cantly greater risk of premature coronary artery disease in FED compared to FLD. This nding is important, it suggests that management should be tailored according to the LCAT de ciency pro le. In FLD patients, the priority is to mitigate progression to end stage kidney disease; in contrast, in FED patients, management of cardiovascular risk may well be paramount. Finally, the LCAT mutations discussed in this article are the only ones reported in the Mexican-Amerindian population. We report a novel mutation associated with FLD, in a Mexico-Mestizo woman, suggesting the in uence of Amerindian ancestry. Competing interests: authors do not have any con icts of interest to disclose in relation to this manuscript.

Abbreviations
Funding: no funding was received for this work.
Authors' contributions: RM, DEL, AJM: conceptualization, methodology, research, analysis, writing, review and edition of the manuscript. OAPM, MLOS, YS: research, analysis, review and edition of the manuscript. MTT, CAAS: conceptualization, methodology, supervision, project administration, funding acquisition, review and edition of the manuscript. All authors read and approved the nal manuscript.