Global disparities in mortality and liver transplantation in hospitalised patients with cirrhosis: a prospective cohort study for the CLEARED Consortium

BACKGROUND

Cirrhosis, the end result of liver injury, has high mortality globally. The effect of country-level income on mortality from cirrhosis is unclear. We aimed to assess predictors of death in inpatients with cirrhosis using a global consortium focusing on cirrhosis-related and access-related variables.

METHODS

In this prospective observational cohort study, the CLEARED Consortium followed up inpatients with cirrhosis at 90 tertiary care hospitals in 25 countries across six continents. Consecutive patients older than 18 years who were admitted non-electively, without COVID-19 or advanced hepatocellular carcinoma, were enrolled. We ensured equitable participation by limiting enrolment to a maximum of 50 patients per site. Data were collected from patients and their medical records, and included demographic characteristics; country; disease severity (MELD-Na score); cirrhosis cause; medications used; reasons for admission; transplantation listing; cirrhosis-related history in the past 6 months; and clinical course and management while hospitalised and for 30 days post discharge. Primary outcomes were death and receipt of liver transplant during index hospitalisation or within 30 days post discharge. Sites were surveyed regarding availability of and access to diagnostic and treatment services. Outcomes were compared by country income level of participating sites, defined according to World Bank income classifications (high-income countries [HICs], upper-middle-income countries [UMICs], and low-income or lower-middle-income countries [LICs or LMICs]). Multivariable models controlling for demographic variables, disease cause, and disease severity were used to analyse the odds of each outcome associated with variables of interest.

FINDINGS

Patients were recruited between Nov 5, 2021, and Aug 31, 2022. Complete inpatient data were obtained for 3884 patients (mean age 55·9 years [SD 13·3]; 2493 (64·2%) men and 1391 (35·8%) women; 1413 [36·4%] from HICs, 1757 [45·2%] from UMICs, and 714 [18·4%] from LICs or LMICs), with 410 lost to follow-up within 30 days after hospital discharge. The number of patients who died while hospitalised was 110 (7·8%) of 1413 in HICs, 182 (10·4%) of 1757 in UMICs, and 158 (22·1%) of 714 in LICs and LMICs (p<0·0001), and within 30 days post discharge these values were 179 (14·4%) of 1244 in HICs, 267 (17·2%) of 1556 in UMICs, and 204 (30·3%) of 674 in LICs and LMICs (p<0·0001). Compared with patients from HICs, increased risk of death during hospitalisation was found for patients from UMICs (adjusted odds ratio [aOR] 2·14 [95% CI 1·61-2·84]) and from LICs or LMICs (2·54 [1·82-3·54]), in addition to increased risk of death within 30 days post discharge (1·95 [1·44-2·65] in UMICs and 1·84 [1·24-2·72] in LICs or LMICs). Receipt of a liver transplant was recorded in 59 (4·2%) of 1413 patients from HICs, 28 (1·6%) of 1757 from UMICs (aOR 0·41 [95% CI 0·24-0·69] vs HICs), and 14 (2·0%) of 714 from LICs and LMICs (0·21 [0·10-0·41] vs HICs) during index hospitalisation (p<0·0001), and in 105 (9·2%) of 1137 patients from HICs, 55 (4·0%) of 1372 from UMICs (0·58 [0·39-0·85] vs HICs), and 16 (3·1%) of 509 from LICs or LMICs (0·21 [0·11-0·40] vs HICs) by 30 days post discharge (p<0·0001). Site survey results showed that access to important medications (rifaximin, albumin, and terlipressin) and interventions (emergency endoscopy, liver transplantation, intensive care, and palliative care) varied geographically.

INTERPRETATION

Inpatients with cirrhosis in LICs, LMICs, or UMICs have significantly higher mortality than inpatients in HICs independent of medical risk factors, and this might be due to disparities in access to essential diagnostic and treatment services. These results should encourage researchers and policy makers to consider access to services and medications when evaluating cirrhosis-related outcomes.

FUNDING

National Institutes of Health and US Department of Veterans Affairs.

Infection and Fumonisin Production by Fusarium verticillioides in Developing Maize Kernels

ABSTRACT Fusarium ear rot and fumonisin contamination are serious problems for maize growers, particularly in the southeastern United States. The lack of maize genotypes highly resistant to infection by Fusarium verticillioides or to fumonisin contamination emphasizes the need for management strategies to prevent contamination by this mycotoxin. Information on the initial appearance of infection and fumonisin contamination of kernels and their increase over time is needed to determine if early harvest may be an appropriate control strategy. Maize ears from replicated studies at two locations in eastern North Carolina were harvested weekly, starting 2 weeks after pollination and continuing for 14 weeks. The percentage of kernels infected with F. verticillioides and the fumonisin contamination in the harvested samples were determined. Kernel infection by F. verticillioides and fumonisin contamination appeared as kernels neared physiological maturity and increased up to the average harvest date for maize in North Carolina. Beyond this date, the concentrations of fumonisin fluctuated. Under years conducive for fumonisin contamination, early harvest (greater than 25% grain moisture) may help reduce the level of contamination.

Timing of phosphorus fertilizer application within an irrigation cycle for perennial pasture

Irrigated pastures are significant contributors of phosphorus (P) to inland watercourses, with much of the P coming from applied fertilizer. It was hypothesized that the timing of P fertilizer application relative to irrigation regulates P concentrations in runoff and infiltrating water. To test this hypothesis, a two-by-two factorial experiment was conducted on twelve 8- x 30-m border-irrigated bays growing perennial pasture. Phosphorus fertilizer in the form of single superphosphate (44 kg P ha(-1)) was surface-broadcast onto the bays when the nominal change in soil water deficit reached 0 or 50 mm (U.S. Class A pan evaporation minus rainfall). Following fertilizer application, the bays were again irrigated when the nominal soil water deficit between fertilizing and the subsequent irrigation reached either 0 or 50 mm. The volume of water applied, runoff volume, and changes in soil water content were recorded for the three irrigations following fertilizer application. Total phosphorus (TP) and filtrable reactive phosphorus (FRP, <0.45 microm) concentrations in runoff and at depths of 0.1, 0.3, and 0.6 m in the soil were also measured. Soil water content at fertilizer application had less effect on P concentrations in runoff and soil water than the additional time between fertilizing and irrigating. By allowing a deficit of 50 mm between fertilizer application and irrigation, the average concentration of P in runoff and moving below a soil depth of 0.1 m was approximately halved. To maximize fertilizer use efficiency and minimize environmental effects, a delay should occur between applying P fertilizer and irrigating perennial pasture.