1. Demographic and Clinical Characteristics
Sixty patients were separated into two groups—ascites mNGS positive group and mNGS negative group—based on whether the ascites mNGS results showed the presence of pathogens. The ascites mNGS positive group consisted of 24 patients. All of them were 55.38 ± 8.67 years old, with 15 males accounting for 62.5% and 9 females accounting for 37.5%,7 patients with hepatitis B cirrhosis (29.2%), 2 patients with hepatitis C cirrhosis (8.3%), 8 patients with alcoholic cirrhosis (33.3%),3 patients with primary biliary cholangitis (PBC) (12.5%), 3 patients with unexplained liver cirrhosis (12.5%), and 1 patient with hepatitis B cirrhosis combined with alcoholic cirrhosis (4.2%). The ascites mNGS-negative group consisted of 36 patients. All of them were 51.92 ± 9.74 years old, with 29 males (80.6%) and 7 females (19.4%). 13 patients with hepatitis B cirrhosis (36.1%), 2 patients with hepatitis C cirrhosis (5.6%), and 14 patients with alcoholic cirrhosis (38.9%), 2 patients with PBC (5.6%), 2 patients with unexplained liver cirrhosis (5.6%), and 3 patients with hepatitis B cirrhosis combined with alcoholic cirrhosis (8%). In terms of the ratio of age, gender, and cirrhosis etiology, there were no significant variations between the two groups (P > 0.05) (Table 1).
Prothrombin (P = 0.005), international standardized ratio (P = 0.014), procalcitonin (P = 0.018), and c-reactive protein (P = 0.042) were significantly higher in the mNGS-positive group than in the ascites mNGS-negative group. In contrast, total serum protein (P = 0.032) and Na+ (P = 0.024) were significantly lower (Table 1). SBP patients with cirrhosis have impaired liver function and infection. PCT and CRP data were missing in certain patients. Only patients having complete PCT and CRP data in both groups (n = 40) were compared in this study, and PCT and CRP in the ascites mNGS positive group were significantly greater than those in the mNGS negative ascites group (P < 0.05). (Table 1).PCT has been evaluated as a biomarker of SBP in cirrhosis in multiple studies, and the results support PCT as a biomarker of SBP in cirrhosis [17–19]. The mNGS-positive group had substantially higher Child-Pugh scores (11 vs. 9.5, P = 0.011), MELD scores (18 vs. 13, P = 0.020), and MELD-Na scores (23.79 vs. 14.60, P = 0.004) than the mNGS-negative group. As a result, cirrhotic patients with a positive mNGS result have more serious liver injury and a worse prognosis. In a single study of each assay index, there was no significant difference between the two groups. However, there was a significant difference in liver disease scores(Child-Pugh scores, MELD scores, and MELD-Na scores) between the two groups. This result shows that conventional laboratory examination had limited sensitivity and had little importance in predicting the prognosis of cirrhotic patients.
Table 1
Demographic and clinical characteristics of the 60 patients
| Total (n = 6) | mNGS positive group (n = 24) | mNGS negative group (n = 36) | P-value |
Age(years) | 53.3 ± 9.41 | 55.38 ± 8.67 | 51.92 ± 9.74 | 0.145 |
Sex | | | | 0.121 |
Male(n,%) | 44(73.3) | 15(62.5) | 29(80.6) | |
Etiology of cirrhosis | | | | 0.552 |
HBV (n, %) | 20(33.3) | 7(29.2) | 13(36.1) | |
HCV (n, %) | 4(6.7) | 2(8.3) | 2(5.6) | |
Alcohol (n, %) | 22(36.7) | 8(33.3) | 14(38.9) | |
PBC (n, %) | 5(8.3) | 3(12.5) | 2(5.6) | |
Unknown factors (n, %) | 5(8.3) | 3(12.5) | 2(5.6) | |
HBV + alcohol (n, %) | 4(6.7) | 1(4.2) | 3(8.3) | |
Blood routine | | | | |
WBC (×109) | 5.36(3.41,8.70) | 7.80(3.20,10.95) | 5.02(3.50,7.16) | 0.105 |
NE (×109) | 3.51(2.07,6.39) | 5.68(2.17,8.70) | 3.08(1.95,4.81) | 0.059 |
LY (×109) | 0.99 ± 0.59 | 0.95 ± 0.62 | 1.02 ± 0.57 | 0.655 |
NLR | 4.20(2.43,6.49) | 13.08(5.34,23.52) | 5.83(3.83,8.24) | 0.072 |
Hb(g/L) | 98.85 ± 25.39 | 98.25 ± 20.30 | 99.25 ± 28.55 | 0.882 |
PLT (×109) | 87.6 ± 48.22 | 93.91 ± 52.92 | 83.38 ± 45.10 | 0.412 |
Liver function test | | | | |
ALT(U/L) | 37.45(18.35,97.25) | 73.25(20.25,150.02) | 29.45(17.62,51.17) | 0.263 |
AST(U/L) | 59.55(32.65,141.97) | 84.75(34.62,205.82) | 57.35(31.52,108.92) | 0.096 |
TBIL (umol/L) | 62.9(36.25,143.77) | 76.90(43.02,402.97) | 51.80(28.85,129.60) | 0.115 |
TP(g/L) | 59.45(53.92,66.52) | 56.4(52.02,63.15) | 61.25(56.57,68.65) | 0.032 |
ALB(g/L) | 27.38 ± 4.62 | 26.73 ± 4.49 | 27.81 ± 4.73 | 0.378 |
Renal function test | | | | |
Scr(umol/L) | 67.7(54.82,91.07) | 68.85(55.72,95.05) | 66.85(51.07,91.05) | 0.460 |
BUN (umol/L) | 5.92(4.05,7.87) | 6.02(4.53,8.71) | 5.37(3.98,7.78) | 0.285 |
Na+(mmol/L) | 133.6(131,137.1) | 132.45(128.2,134.9) | 135.4(131.5,137.6) | 0.024 |
Coagulation test | | | | |
PT(s) | 16.3(14.7,18.97) | 18.7(15.25,22.92) | 15.35(14.62,17.47) | 0.005 |
PTA | 54.85% ± 14.97% | 50.66% ± 17.75% | 57.65% ± 12.27% | 0.102 |
FIB(g/L) | 1.72(1.22,2.28) | 1.43(1.04,2.15) | 1.88(1.43,2.35) | 0.095 |
INR | 1.42(1.25,1.66) | 1.58(1.27,1.96) | 1.34(1.24,1.53) | 0.014 |
Ascites indicators | | | | |
WBC (n/mm3) | 169(91.75,245.5) | 180(65.75,245.5) | 161(114.25,247.75) | 0.860 |
PMN (n/mm3) | 17.15(10.24,38.34) | 17.15(9.4,40.63) | 17.20(10.68,37.13) | 0.680 |
Total protein(g/L) | 8.81(6.61,14.58) | 7.82(4.52,14.17) | 8.98(7.55,14.86) | 0.228 |
Infection indicators | | | | |
PCT# | 0.48(0.19,1.12) | 0.67(0.41,2.83) | 0.31(0.17,0.64) | 0.018 |
CRP# | 18.58(12.24,37.26) | 25.93(15.47,56.67) | 13.36(10.51,35.65) | 0.042 |
Scores | | | | |
Child-Pugh Score | 11(9,12) | 11(10,12.75) | 9.50(8.25,12) | 0.011 |
MELD Score | 14(10,20.25) | 18(11,24.25) | 13(9.25,15.75) | 0.02 |
MELD-Na Score | 16.25(8.55,26.36) | 20.87(15.98,30.93) | 11.66(7,19.50) | 0.004 |
HBV, hepatitis B virus; HCV, hepatitis C virus; PBC, primary biliary cholangitis; WBC, white blood cell; NE, neutrophil; LY, lymphocyte; NLR, neutrophil to lymphocyte ratio; Hb, hemoglobin; PLT, platelet; ALT, alanine aminotransferase; AST, aspartate aminotransferase; TBIL, total bilirubin; TP, total protein; ALB, albumin; Scr, serum creatinine; BUN, blood urea nitrogen, PT, prothrombin; PTA, prothrombin activity; FIB, fibrinogen; INR, international normalized ratio; PMN, polymorphonuclear leukocyte. PCT, procalcitonin; CRP, c-reactive protein. P values were obtained using the Student's t-test or the Wilcoxon signed-rank test. #: the number of PCT test in mNGS positive group is 17 while number of PCT test in mNGS negative group is 23; number of CRP test in mNGS positive group is 17 while number of CRP test in mNGS negative group is 23.
2. The Distribution Of Pathogens Detected By Mngs And Ascites Culture
A total of 50 strains of positive pathogens were detected by mNGS in 24 patients, with 18 G+ bacteria accounting for 36%, 22 G− bacteria accounting for 44%, 4 fungus accounting for 8%, and 6 viruses accounting for 12% (including parvovirus, Epstein-Barr virus, human herpes virus, and hepatitis B virus).Staphylococcus was the most common G+ bacteria, whereas Escherichia coli and acinetobacter were the most common G− bacteria. 14 strains of positive pathogens were detected by ascites culture, with 9 G+ bacteria accounting for 64.3%, 4 G− bacteria accounting for 28.6%, and 1 fungi accounting for 7.1%.
The detailed pathogens information on ascites of the subjects is shown in Table 2. The total positive rate of pathogens detected by the mNGS method was higher than conventional culture, and more pathogens and more detailed classifications were detected. The mNGS method can obtain the number of pathogen sequences, guiding clinical treatment more comprehensively. The mNGS method was used to obtain the pathogens’ results faster than ascites culture, but this study did not record the specific time. Interestingly, 4 patients who died during the follow-up period had viral (Torque Teno virus, EBV, Human gamma herpesvirus, Human herpesvirus beta 5) ascites by the mNGS detection. The mNGS result of ascites in another survivor contained HBV. Some patients choose to leave the hospital for their reasons/economic reasons and do not continue to receive treatment.
Three pathogens, including one bacterium and two viruses, were found in the mNGS results of patients with number 1. Acinetobacter baumannii was the only bacterium found in ascites culture. Three pathogens, including Enterococcus lead flavococcus, Aspergillus, and Streptomyces arborifolia, were identified by mNGS in patients with number 4, however only coagulase-negative staphylococcus was identified by ascites culture. In patients with number 10, the mNGS method identified Escherichia coli, Escherichia fergusson, Shigella baumannii, and Shigella dysentery; however, only Escherichia coli was identified in the ascites culture. Patients with the numbers 2, 5, 12, and 19 died during the 28-day follow-up period, while those with the numbers 13 and 20 died during the 90-day follow-up period. (Table 2).
Table 2
Pathogens information of subjects in the mNGS detection and ascites bacterial culture
Patients ID | mNGS result(Sequence number) | Results of Ascites culture | PMN(n/mm3) |
1 | Corynebacterium striatum(441)、Torque teno virus(18)、Leptocyclic virus type 16(61) | Acinetobacter baumannii | 119.18 |
2 | Staphylococcus petenkov(50)、Streptococcus mitis(5)、Prevotella amniotic fluid(28) | - | 16.83 |
3 | Staphylococcus hominis(33)、Staphylococcus epidermidis(25) G + streptococcus(7) | - | 9.12 |
4 | Enterococcus casselifavus (9)、Aspergillus(21)、Alternaria arboreal(18) | Coagulase-negative staphylococcus | 15.66 |
5 | Klebsiella pneumoniae(7) | - | 323.31 |
6 | G+ streptococcus(12) | - | 10.24 |
7 | G+ Cook fungus of marsh(12) | - | 23.92 |
8 | Stenotrophomonas rhizophila(6)、Escherichia coli(1) | - | 43.05 |
9 | Staphylococcus epidermidis(31) | Staphylococcus epidermidis、Viridans streptococci | 10.64 |
10 | Escherichia coli(130)、Escherichia fergusonii(3)、Shigella bogdii(11)、Shigella dysenteriae(5) | Escherichia coli | 244.62 |
11 | Debali yeast(11) | - | 17.22 |
12 | Streptococcus pneumonia(70), Streptococcus mitis(6) | - | 17.08 |
13 | Staphylococcus aureus(19), EBV(78) | - | 19.1 |
14 | Klebsiella heteroliticus(4) | Micrococcus luteus | 8.82 |
15 | Staphylococcus petenkov(70)、Staphylococcus warneri(18) | - | 14.82 |
16 | Acinetobacter pittobacter(24493)、Acinetobacter(1456)、Enterobacter asburiae(2375)、Enteric bacilli(7)、Citrobacter flodi(7) | Acinetobacter baumannii、Enterococcus faecalis、Bacillus cereus | 33.39 |
17 | Enterococcus casselifavus(223)、Enterococcus pallidans(93) | Staphylococcus capitis | 5.52 |
18 | Aspergillus(149)、G− Pre-votella multiformis(219)、 Pre-votella nigrescens(40) | - | 5.68 |
19 | Human herpesvirus beta 5(2) | Candida albicans | 19.2 |
20 | Human gamma herpesvirus type 4(6) | - | 1038.42 |
21 | Staphylococcus hominis(49)、Staphylococcus epidermidis(13)、Prevotella(7)、Shewanella oneidensis(7) | - | 5.28 |
22 | Pseudomonas aeruginosa(4)、Klebsiella pneumoniae (2) | Difference of veillon coccus | 5.9 |
23 | Hepatitis B virus(15) | - | 2159.7 |
24 | Enterobacter cloacae(1) | - | 29.88 |
25 | - | Viridans streptococci | 24.9 |
26 | - | Staphylococcus warneri | 12.2 |
The mNGS and ascites culture results of subjects not listed in the table were negative
3. The Value Of Ascites Mngs Results In The Diagnosis Of Sbp In Cirrhotic Patients
The classification of the 60 patients with cirrhotic ascites included in this study was based on ascites mNGS. 25 patients were diagnosed with SBP (41.67%) when PMN was combined with ascites mNGS detection, including 3 patients with typical SBP, 1 patient with culture-negative neutrophilic ascites (CNNA), 21 patients with bacterial ascites. At the same time, cirrhotic patients were classified based on ascites culture results. A total of 15 patients (25%) were diagnosed with SBP (41.67%) when PMN was combined with ascites culture, 4 patients with CNNA, 11 patients with bacterial ascites. Typical SBP could not be diagnosed by ascites culture (Table 3). The positive rate of ascites PMN combined with ascites mNGS detection in the diagnosis of SBP in cirrhotic patients (41.67%) was significantly higher (P < 0.05) than that of ascites PMN combined with ascites culture (25%).
Table 3
Ascites classification of subjects by mNGS method and culture method
| mNGS method (n = 60) | Ascites culture method (n = 60) | P-value |
SBP | | | |
SBP (n; %) | 3(5) | 0(0) | |
CNNA (n; %) | 1(1.67) | 4(6.67) | |
BA (n; %) | 21(35) | 11(18.33) | |
Total (n; %) | 25(41.67) | 15(25) | < 0.001 |
Non-SBP | | | |
AA (n; %) | 35(58.33) | 45(75) | |
SBP, spontaneous bacterial peritonitis; CNNA, culture-negative neutrophilic ascites; BA, bacterial Ascites; AA, aseptic ascites. Pvalues were obtained using the chi-square test.
4. The Value Of Ascites Mngs Results In The Prognosis Of Cirrhotic Ascites Patients
We prospectively collected the prognostic information of 60 included patients at 28-day and 90-day after enrollment to further understand the connection between ascites mNGS results and the prognosis of cirrhotic patients with ascites. At 28-day and 90-day after enrollment, patients in the mNGS-positive group had a significantly higher mortality rate than those in the mNGS-negative group (16.7% vs. 0, P = 0.022 at 28-day after enrollment; 29.2% vs. 2.7%, P = 0.003 at 90-day after enrollment) (Fig. 1). The 90-day mortality rate did not differ significantly between patients in the ascites culture-positive and the ascites culture-negative groups. The results of ascites culture had little influence on the judgment of the prognosis of cirrhotic patients with ascites in a short time. In other words, patients with cirrhotic SBP have higher short-term mortality than non-SBP.
All of the subjects in this study were acute decompensation (AD) of cirrhosis as they were cirrhotic patients with ascites at enrollment. The proportion of pre-ACLF patients who developed ACLF during the 28-day follow-up period in the ascites mNGS-positive group was higher than that in the ascites mNGS-negative group (50% vs. 5.6%, P < 0.001). The proportion of pre-ACLF patients in the ascites mNGS-positive group was considerably more significant than that in the ascites mNGS-negative group during the 90-day follow-up period (54% vs. 11.1%, P < 0.001) (Table 4). The results show that patients in the mNGS positive group have more severe diseases and have a worse prognosis than those in the mNGS negative group, which is consistent with those patients in the mNGS positive group who have a more significant density of events and mortality. The mNGS approach for detecting ascites, which can improve the diagnosis rate of SBP in cirrhosis, is mainly responsible for the above results.
Table 4
Distribution of clinical courses in subjects during the follow-up period
| mNGS positive group (n = 24) | mNGS negative group (n = 36) | P-value |
28-day follow-up period | | | 0.002 |
pre-ACLF (n; %) | 12(50) | 2(5.6) | < 0.001 |
UDC (n; %) | 5(21) | 7(19.4) | 0.895 |
SDC (n; %) | 7(29) | 27(75) | < 0.001 |
90-day follow-up period | | | 0.002 |
pre-ACLF (n; %) | 13(54) | 4(11.1) | < 0.001 |
UDC (n; %) | 5(21) | 13(36.1) | 0.206 |
SDC (n; %) | 6(25) | 19(52.8) | 0.033 |
ACLF, acute-on-chronic liver failure; SDC, stable decompensated cirrhosis; UDC, unstable decompensated cirrhosis. P values were obtained using the chi-square test.