Basic characteristics of the study participants are presented in Table 1.
Table 1: Clinical and biochemical characteristics.
|
Healthy persons
(n = 10)
|
Patients with cirrhosis
(n = 10)
|
Age (years)
|
39 (22–66)
|
58 (43–73)
|
Body weight (kg)
|
85 (79–91)
|
77 (67–87)
|
Total muscle mass (kg)
|
31 (29–34)
|
26 (21–31)
|
Child-Pugh score
|
-
|
7 (6–9)
|
MELD-Na score
|
-
|
12 (10–14)
|
Ascites (0/1/2/3)
|
-
|
8/0/2/0
|
Biochemistry
|
Alanine aminotransferase (U/L)
|
20 (14–26)
|
33 (23–43)
|
Bilirubin (μmol/L)
|
12 (8–17)
|
23 (15–31)
|
Albumin (g/L)
|
40 (38–42)
|
28 (23–34)
|
Creatinine (μmol/L)
|
75 (67–84)
|
66 (55–77)
|
Haemoglobin (mmol/L)
|
8.5 (8.3–8.7)
|
7.1 (6.0–8.2)
|
Thrombocytes (x109/L)
|
236 (195–277)
|
87 (58–116)
|
MELD-Na, model for end-stage liver disease-sodium. Values are given as means with their 95% CIs.
Baseline arterial ammonia was lower in healthy persons as compared with the patients with cirrhosis (14 [10–19] μmol/L vs. 53 [32–74] μmol/L, p < 0.001, Table 2). Femoral vein concentrations were significantly lower in both groups (p ≤ 0.03) [8 (4–12) and 40 (23–57) μmol/L, respectively, < 0.001). The arterial-venous ammonia concentration gap did not differ significantly between the groups (p = 0.15). During ammonia infusion at t = 80 minutes, ammonia concentrations in the radial artery and the cubital and femoral veins were 115 (97–133), 61 (31–90), 47 (32–59) μmol/L, respectively, in healthy persons and 175 (123–227), 134 (65–203), 120 (84–155) μmol/L, respectively, in the patients with cirrhosis (Table 2).
Table 2 Ammonia concentrations before and after induced hyperammonaemia
|
Healthy persons
(n=10)
|
Patients with cirrhosis
(n=10)
|
ULN values
(for patients with cirrhosis)
|
Arterial, baseline
|
14 (10–19)
|
53 (32–74)
|
2.0 (1.2–2.8)
|
Femoral venous, baseline
|
8 (4–12)
|
40 (23–57)
|
2.1 (1.2–3.0)
|
Arterial, infusion
|
115 (97–133)
|
175 (123–227)
|
6.7 (4.7–8.7)
|
Femoral venous, infusion
|
47 (32–59)
|
120 (84–155)
|
6.2 (4.4– 8.1)
|
Cubital venous, infusion
|
61 (31–90)
|
134 (65–203)
|
-
|
Mean ammonia concentrations for healthy persons and patients with cirrhosis (μmol/L) and calculated upper limit normal (ULN) values for patients with cirrhosis with their respective 95% CIs
Data collected after 80 minutes of ammonia infusion were analysed by multivariate model testing showing no significant interaction between the effects of sampling site and cirrhosis (p = 0.75). Thus, the effect of sampling site were similar for healthy persons and cirrhosis patients during induced hyperammonaemia. Therefore, an additive model was chosen accepting parallel slopes between the two groups (Figure 1). The average ammonia concentration difference between healthy persons and patients with cirrhosis across all three sampling sites was 72 (42–103) μmol/L p < 0.001. Table 3 presents ammonia concentration gaps between the radial artery, the cubital vein and the femoral vein. During induced hyperammonaemia, ammonia concentrations were non-significantly higher in the cubital vein as compared with the femoral vein (10 [-8–29] μmol/L, p = 0.26). In both groups, arterial concentrations were ~ 50 μmol/L (p < 0.001) higher than venous concentrations (Table 3).
Table 3 Ammonia concentration differences between radial artery, cubital vein and femoral vein during induced hyperammonaemia
Sampling site comparison
|
Δ [Ammonia]
|
p
|
a. radialis vs. v. cubitalis
|
50 (37–64)
|
<0.001
|
a. radialis vs. v. femoralis
|
61 (46–76)
|
<0.001
|
v. cubitalis vs. v. femoralis
|
10 (-8–29)
|
0.26
|
Mean ammonia concentration differences with their 95% CIs (Δ, μmol/L)
In the patients with cirrhosis, ammonia concentrations normalised to ULN yielded comparable values for arterial and venous blood at baseline [2.0 (1.2–2.8) and 2.1 (1.2–3.0) times ULN, respectively, p = 0.74] and during induced hyperammonaemia [6.7 (4.7–8.7) and 6.2 (4.4–8.1) times ULN, respectively, p = 0.44] (Table 2).