Cirrhosis is the end-stage of different etiology-related chronic liver diseases, being alcohol-related (AR), hepatitis C virus infection (HCV) and non-alcoholic steatohepatitis (NASH) the most commonly involved [1,2].

Along its natural history the disease undergoes different progressive stages, either considered as compensated or decompensated [3,4]. The compensated state, usually the longest, is almost devoid of symptoms and therefore, liver disease related hospital admissions are uncommon. On the other hand, almost every event characterizing the decompensated stage requires admission, in order to be properly diagnosed and treated.

Main decompensating events are: esophageal variceal bleeding (EVB), ascites, hepatic encephalopathy (HE) and jaundice. Closely related, either as the cause or as a consequence, infections and hepatorenal syndrome (HRS) are frequently present in decompensated patients [4–7]. Alone or combined, all these complications represent a threatening condition in cirrhotic patients, decreasing their life expectancy [4,8,9]. Moreover, the growing incidence of primary liver cancer, mostly hepatocellular carcinoma (HCC), is another determinant of frequent nosocomial admissions [10].

For each of these complications, diagnostic and therapeutic algorithms have been proposed and translated into different valid guidelines [11–16]. The challenging consequence of these recommendations is a permanent requirement of both adequate medical and technical resources at the institutions where these patients are treated. In fact, decompensated cirrhotic patients are mainly admitted to Gastroenterology or Int Med Depts, as Liver Units are usually restricted to specific referral centers, mostly located at great urban areas. Given the diversity of institutions involved, a potential lack of homogeneity may translate into uneven quality in medical care provided for those decompensating events [17–19].

Therefore, the present study was designed in order to search: (a) the demographic and clinical–epidemiological characteristics of cirrhotic patients hospitalized on certain date at different medical centers of Central and South America and (b) the availability of resources and expertise for diagnosis and treatment of the most frequent complications of liver cirrhosis.

A cross-sectional, multicenter, observational study, performed on March 21st, 2018. Every national society affiliated to the Latin American Association for the Study of the Liver (ALEH) and all regional associations subsidiaries of the Argentine Federation of Gastroenterology (FAGE) were invited to partipate. Those accepting provided a list of medical institutions where cirrhotic patients are frequently admitted. Invitations including detailed objectives and design of the study were mailed. Those centers accepting completed the questionaire with data corresponding to the aforementioned date.

The whole survey contained 52 questions, divided in 3 sections: (1) data related to the participating medical center and its location (10 questions), (2) a clinical/demographic profile of every cirrhotic patient remaining hospitalized on March 21st/2018 (13 questions) and (3) an inquiry regarding usual therapeutic attitudes and corresponding available resources for the management of (a) EVB; (b) ascites, spontaneous bacterial peritonitis (SBP), HRS and non-SBP infections; (c) HE; (d) liver transplantation (LT) and (e) HCC (29 questions).

The investigation was approved by the Human Investigation Committees of participating institutions.

All data were centralized at one center (Gastroenterology and Liver Dept, University of Rosario Medical School, Rosario, Argentina) and assembled and processed by JV, LH and FT.

Cirrhosis is a major cause of morbidity and mortality worldwide [1,2,22–24]. Regardless of successful preventive measures and specific therapies for certain etiologies [25–27], the growing incidence of other agents as main determinants of chronic liver injury sustains its prevalence [2,28]. Advanced cirrhotic patients face the risk of decompensation that, once present, requires frequent hospital admissions [5–7,29–32]. In order to diagnose and treat them adequately, they should take place at centers where both medical know-how and updated technical resources are available. However, this is not always feasible, and patients may be admitted at institutions lacking expertise for their management [17–19].

The present survey, the first of its kind performed in Latin America, analyses data obtained from a group of mostly urban located centers, sharing a high degree of academic affiliation. The existence of Gastroenterology Depts at most of them, frequently associated to Liver Units, explain the significant presence of decompensated cirrhotic patients at these institutions.

The final data includes 13 (81%) of 16 originally invited countries, with a global answer rate of 37%. The contribution was quite heterogeneous, showing, in some specific cases, a disproportion between country population and the number of participating centers. Therefore, 53% of the countries recruited 85% of the centers providing 93% of included patients. Given this limited territorial projection, representative value of these data may be questioned. However, as mentioned before, no significant difference was observed between those patients recruited at major contributing countries (>30 pts) versus those included at less contributors.

AR was the global leading cause. However, as previously mentioned, NASH resulted highly prevalent within patients enrolled in the Pacific coast (Chile, Perú and Ecuador). In spite of this, BMI categories and diabetes incidence did not differ between these 3 countries and the others (not shown in Results). Reasons for this distinctive feature remain unknown as neither etnicity nor dietary habits were part of the inquiry.

Male predominance, a treshold age of 40 years for the majority and both CP/MELD scores reflecting advanced disease characterized this population. Interestingly, a BMI>25 and diabetes were present in 57% (n=214) and 32% (n=122) of them, respectively. AR and NASH were the most prevalent etiologies, both accounting for 63% (n=235) of the population while HCV, the third related cause, represented 7% (n=25) of the cases. This profile is coincident with reports showing a shift toward AR and NASH as the main contemporary emergent agents of chronic liver disease [1,2,23,24,28,33,34]. Interestingly, HCV infection was higher in females, as has been previously reported in specific Latin American geographic areas [35] (Supplementary Table 3).

Remarkably, a low vaccination rate was observed in these patients, far away from recommended schedules for cirrhotic patients [36].

Sign and symptom overlap (i.e. GIB/HE, infections/HRS) is frequent in decompensated cirrhotics, making identification of a single admission cause occasionally difficult. Therefore, given the high frequency of such combinations, admission cause was defined as the prevailing feature at presentation (“main admission cause”). Accordingly, ascites, GIB, HE and infections were the main identifiable admission causes, as reported elsewhere [7,32,37,38].

Regarding EVB, recommended steps are adequately followed at most institutions [11]. Nevertheless, some disparities deserve comment. Bleeders are “always” admitted to ICUs at only 20% of these institutions. According to the Baveno VI consensus meeting “patients with acute variceal hemorrhage should be considered for ICU or other well-monitored units” [11]. Although desirable, wether most centers not admitting bleeders to ICUs are doing it at “other well-monitored units” is difficult to assume. Vasoactive drugs and ATB administration associated to diagnostic/therapeutic endoscopic procedures performance was homogeneously qualified among all centers. A remarkable point is the unavailability of a SBT at 32% (n=21) of the places while, on the other hand, SBT is the only available therapeutic tool at 5 of them. May this implie its under or over usage is hipothetical. However, given the limited availability of both emergency TIPS (8 centers; 12%) and self-expandable metal stents (11 centers; 17%), possession of a SBT must be strongly encouraged.

Ascites therapy is a well standardized issue [13,19], however, care of these patients may be suboptimal [18]. The fact that in the present survey diagnostic paracentesis is performed “only” when SBP is suspected in almost 30% of the centers is worrisome, as SBP may be absolutely asymptomatic. Moreover, its suspicion is not the only indication to perform a diagnostic paracentesis [13,39,40]. This criterion must be necessarily reconsidered. Both standard medical treatment of ascites and therapy at its refractory status were homogeneously satisfactory, endorsing guidelines recommendations [13,40,41]. On the other hand, 1 every 3 centers lacks an adequate treatment for HRS. Finally, regarding SBP, a delicate problem is the low bacterial rescue in ascitic fluid in almost 60% of the institutions, well below reported standard results [41,42]. Inadequate sample collection and/or lack of updated culture technology may explain these dissapointing results.

Prevention of HE either during the acute bleeding episode or as therapy of its chronic stage is accomplished in most centers [15]. Regarding LT, it is available at 32% (n=21) of participating hospitals endorsing the assertion that referral centers were highly represented at this survey. Moreover, liver disease etiology of transplanted patients shows the same changing trend actually observed in the United States [43–45].

Remarkably, HCC screening is performed at 95% (n=62) of the institutions. This elevated screening index is also mirroring a group of mostly liver disease devoted institutions, overcoming suboptimal surveillance observed in non-liver disease oriented centers [46,47]. Nevertheless, a mix of different treatment proposals is only available at 33 of the centers. Finally, medical skills, updated technology and a higher level of adherence to guidelines recommendations prevailed in centers located at the most populated areas (>500,000) as well as in those having simultaneous Gastroenterolgy, Liver and Int Med Depts.

A weakness of this study may be its representative value. Nevertheless, in this group of mainly urban located centers, patient characteristics were homogeneous, regardless number included and countries considered. Therefore, we believe the results are confident enough to accomplish the survey objectives. It may also be argued that Acute on Chronic Liver Disease (ACLF) was not considered among admission causes. In fact, proposal of ACLF as a late decompensation event appeared on a subsequent date of this survey development [4]. Additionally, the survey may be qualified as biased, not acknowledging the “real world”. In fact it is so. As mentioned, medical care of decompensated cirrhotic patients requires a degree of complexity not available anywhere, but at referral centers, to which this survey was intentionally addressed.

In conclusion, this study updates the current status of capacities aimed at the medical care of decompensated cirrhotic patients in a group of Latin American medical centers. Notwithstanding its limited representation, it shows a high degree of resources availability, mostly according with current guidelines recommendations. Several facilities are available, mainly at referral centers, thereby providing the coordinated and multidisciplinary interventions required for the adequate treatment of these patients [18,19,48].AbbreviationsAIH

autoimmune hepatitis