Abstract
Objective
To investigate haemodynamic changes in the transplanted liver without postoperative complications.
Methods
Colour Doppler ultrasound was used to monitor recipients of liver transplants who had no postoperative complications. The haemodynamic data for the hepatic vasculature were compared at different time-points during the first 4 years after liver transplantation.
Results
A total of 144 liver transplant patients were enrolled in the study. Portal vein flow velocity decreased significantly from 72.1 ± 30.3 cm/s at 1 day to 44.2 ± 20.1 cm/s at 1 month after liver transplantation. Hepatic artery flow velocity was 61.4 ± 33.2 cm/s at day 1; it then decreased slowly but significantly to 48.3 ± 20.4 cm/s at 3 years after transplantation. There were 81 (56.3%) patients with high hepatic artery resistance index (HARI) (>0.80) and 19 (13.2%) with low HARI (<0.50) measured at least once during the 4-year follow-up examination.
Conclusions
Decreased portal vein flow velocity was the typical change observed during the first month after liver transplantation. Abnormal haemodynamic Doppler results should be interpreted with caution because they may not be clinically significant and may improve spontaneously.
Introduction
Since the first case of human liver transplantation was reported in 1963, liver transplantation has been widely used to treat end-stage liver disease.1–3 Ultrasound is a convenient and noninvasive approach to follow-up liver transplant recipients. Many authors have reported the value of Doppler ultrasound for the diagnosis of vascular complications after liver transplantation.4–7 However, little is known about the physiological changes during recovery of graft function after liver transplantation. The objective of this study was to investigate the normal hepatic haemodynamic changes in transplanted liver during the recovery of graft function with colour Doppler flow imaging (CDFI).
Patients and methods
Patient population
This retrospective study enrolled consecutive patients who had undergone liver transplantation in the Department of Ultrasound, Zhongshan Hospital, Fudan University, Shanghai, China between October 2005 and March 2009. The study evaluated each liver transplant recipient according to Zhongshan Hospital’s Doppler ultrasonography protocol. Transplant recipients with complications of any kind were excluded from the study. In addition, patients were excluded from the study if there were any anatomical/procedural difficulties in assessing the proper anatomical structures.
The study protocol did not require approval by an ethics committee or institutional review board because ultrasound examinations are noninvasive and they were already being used as part of the routine follow-up examination of liver transplant patients to aid the diagnosis of vascular complications. Verbal informed consent was obtained from all patients enrolled in the study.
Postoperative Doppler ultrasound examinations
Routine, postoperative ultrasound examinations were performed at the following times: once a day during the first 7 days, once a week for the following 2–4 weeks, then once every 3 months for up to 1 year, and then once every 3–6 months for up to 4 years.
Ultrasonography was performed with a commercially available colour Doppler scanner (iU22 xMATRIX ultrasound system; Philips Healthcare, Best, The Netherlands) and a C5-2 Broadband Curved Array Transducer (Philips Healthcare). The hepatic artery resistance index (HARI) was automatically calculated according to the formula: (peak systolic velocity – peak end diastolic velocity)/peak systolic velocity. The Doppler sample volume was maintained in the 1–2 mm range, and the Doppler angle was maintained at <60°.
The following Doppler characteristics of the common hepatic artery and left and right intrahepatic arteries were documented: (i) diameter; (ii) hepatic artery peak systolic velocity (HAPSV, cm/s); and (iii) HARI. The anatomical sites for haemodynamic measurements were as follows: the segment of the common hepatic artery adjacent to the hepatic hilus; the left hepatic artery at the site adjacent to the sagittal portion of the portal vein; and the right hepatic artery within the right anterior arterial branch, 1 cm distal from the bifurcation.
The hepatic portal vein peak velocity (PVPV) and diameter were measured at the anastomotic and nonanastomotic sites (1 cm proximal to the anastomotic point). Hepatic vein peak velocity and diameter were measured in the left, middle and right hepatic veins separately. Furthermore, the blood flow waveforms in all three hepatic veins were analysed. The waveforms were classified into three types: triphasic (two antegrade waves below the baseline and one presystolic retrograde wave above the baseline), biphasic (without the reversed presystolic wave) and monophasic (with a flat pattern). Each hepatic vein was assessed at 3 cm distal from the inferior vena cava.
Statistical analyses
Hepatic blood flow parameters measured in transplanted livers at different postoperative time-points during the first 4 years were analysed to show the haemodynamic changes after liver transplantation. The haemodynamic parameters measured at the 6-month time-point were used for analysing the hepatic vascular haemodynamics after transplantation. Data are presented as mean ± SD. All statistical analyses were performed using the SPSS® statistical package, version 13.0 (SPSS Inc., Chicago, IL, USA) for Windows®. Differences between groups of data were evaluated using analysis of variance and Student’s t-test. Categorical data were compared with χ2-test. A P-value <0.05 was considered statistically significant.
Results
This study enrolled 113 men and 31 women with a mean ± SD age of 48.9 ± 9.2 years (range, 22–72 years). All of the patients received whole grafts and none developed significant graft dysfunction requiring treatment within the 4 years following transplantation (routine liver function tests showed improvement). Pretransplantation diagnoses included 132 cases of viral liver disease, six of alcoholic liver disease, four of cholestatic liver disease, and one each of Wilson’s disease and autoimmune liver disease. In all cases, transplantation was indicated by cirrhosis with (n = 127) or without (n = 17) carcinoma.
The mean ± SD length of the intensive care stay of the 144 patients was 7.1 ± 1.5 days and the mean ± SD length of the hospital stay was 15.7 ± 2.6 days. None of the patients had vascular complications during the 4 years of follow-up examinations with CDFI, such as vascular complications, biliary complications, rejection and tumour recurrence. During the process of recovery, patients had some minor complications including pleural effusion, ascites and transient liver function abnormalities in the early postoperative period.
Figure 1 shows the haemodynamic changes after liver transplantation. Significant hepatic haemodynamic changes occurred within 1 month after transplantation, including a marked reduction in mean ± SD PVPV (72.1 ± 30.3 cm/s at 1 day after transplantation versus 44.2 ± 20.1 cm/s at 1 month; P < 0.05). The mean ± SD PVPV decreased slowly from 44.2 ± 20.1 cm/s at 1-month follow-up to 25.4 ± 8.1 cm/s at 4-year follow-up (P < 0.05). The mean ± SD HAPSV decreased slowly and moderately, from 61.4 ± 33.2 cm/s at 1 day after transplantation to 48.3 ± 20.4 cm/s at 3 years after transplantation (P < 0.05), and increased to 50.5 ± 14.3 cm/s at 4 years (P < 0.05).
Postoperative changes in portal venous peak velocity (PVPV; diamonds) and hepatic artery peak systolic velocity (HAPSV; squares) in patients who had undergone liver transplantation (n = 144). Parameters were recorded at eight time-points after liver transplantation. Data presented as mean ± SD. The colour version of this figure is available at: http://imr.sagepub.com.
Hepatic vascular haemodynamics measured at the 6-month postoperative time-point in patients who had undergone liver transplantation (n = 144) without complications.
Data presented as mean ± SD.
P < 0.05 for overall comparisons among CHA, RHA and LHA; analysis of variance.
P < 0.05 for comparison between PVA and PVPA; Student’s t-test.
P < 0.05 for overall comparisons among LHV, MHV and RHV; analysis of variance.
CHA, common hepatic artery; RHA, right hepatic artery; LHA, left hepatic artery; PVA, portal vein at the site of the anastomosis; PVPA, portal vein at the site proximal to the anastomosis; LHV, left hepatic vein; MHV, middle hepatic vein; RHV, right hepatic vein.
The HARI ranged from 0.4 to 1.0 for the entire study population during the 4-year follow-up ultrasound examinations. Normal resistance was defined as a HARI within the range 0.5–0.8. A total of 81 of the 144 (56.3%) transplant recipients had high resistance (HARI > 0.8) at least once during the follow-up period. HARI in 74 of these 81 (91.4%) patients recovered to the normal range within 6–12 months after liver transplantation. In four of the remaining patients, arterial resistance remained high up to 4 years after transplantation. In contrast, in 19 of the 144 (13.2%) recipients, HARI was <0.5 at least once during the follow-up period but normalized within 6 months. Abnormal HARI was easier to record within 1 year after liver transplantation than at any other time during the 4-year follow-up period.
Portal venous anastomosis of transplanted livers could be clearly recognized as enhanced punctate echoes on the anterior and posterior vascular walls (Figure 2). Portal venous haemodynamic data for the anastomotic and nonanastomotic sites showed that the inner diameter of the portal vein at the anastomotic site was significantly smaller than at the nonanastomotic site (P < 0.05) (Table 1). The portal venous peak velocity at the anastomotic site was slightly faster than at the nonanastomotic site, but the difference was not statistically significant.
A representative ultrasonogram of the intercostal plane that shows punctate echoes on the anterior and posterior walls of the portal vein (PV) at the portal anastomosis (arrow) of a transplanted liver.
Doppler waveforms of the three hepatic veins measured at the 6-month postoperative time-point in patients who had undergone liver transplantation (n = 144) without complications.
Data presented as n (%).
LHV, left hepatic vein; MHV, middle hepatic vein; RHV, right hepatic vein.
No statistically significant between-group differences (P ≥ 0.05).
Discussion
The current sample of 144 liver transplant recipients was larger than samples from previous studies.8,9 The present study excluded recipients with major complications because vascular blood flow could be affected by such complications. In the present study, a total of 21 patients had a peak hepatic arterial velocity >100 cm/s, indicating that detection of a high flow velocity in an isolated arterial segment may not indicate arterial stenosis. Following anastomosis of the transplanted hepatic artery, there is often some redundancy of the hepatic artery, which may form a kink and narrow the arterial lumen without a decrease in arterial wall compliance, resulting in accelerated flow velocity. 10 In addition, accurate determination of peak velocity requires correct determination of the angle at which the segment of the vessel being evaluated lies relative to the skin surface. Colour Doppler has poorer spatial resolution than B-mode imaging, so it is possible that kinks in the hepatic artery went undetected, and spuriously high readings might have occurred.11,12 In this present series, the arterial velocity was significantly greater in the common hepatic artery segment than in the left or right intrahepatic artery segments. Therefore, the site for monitoring hepatic artery blood flow should also be considered.
In the present study, the most prominent haemodynamic change after liver transplantation was a markedly decreased portal venous peak velocity. Sugimoto et al. 13 demonstrated that portal venous velocity dramatically increased immediately after transplantation in patients with cirrhosis compared with those without cirrhosis. This may be related to long-standing splanchnic hyperaemia due to portal hypertension before surgery. 14 In this present study, all patients undergoing transplantation were indicated by cirrhosis with or without a malignant liver mass. The portal vein velocity was high at 1 day after liver transplantation, then decreased gradually, recovering to a normal level up to 4 years after liver transplantation, whereas a rapid decline occurred in the first postoperative month.
The HARI the most commonly used Doppler parameter in hepatic artery evaluation. Normal values, both in healthy individuals and those with transplants, range from 0.50 to 0.80. 9 In this current series, a high-resistance hepatic arterial flow (HARI >0.80) was a frequent finding. High HARI has not been associated with worsening of clinical course or graft function; it may be related to older donor age and prolonged cold ischaemia. 15 Furthermore, this present study also found that a high HARI was more frequent during the early postoperative period. This was probably a result of the hepatic buffer response. This is a mechanism that tends to maintain a constant hepatic blood flow by decreasing arterial blood flow when portal vein blood flow dramatically increases.16–18 In addition, a decreased HARI (<0.50) can also be observed in the early postoperative period, although it was less common than a high HARI. These low HARI values might be caused by surgical oedema, with low HARI values ordinarily returning to a normal level within 6 months. Low HARI can also be observed in other pathological conditions, including hepatic artery stenosis, severe aortoceliac atherosclerotic disease, arteriovenous or arterial–biliary fistula formation, and hepatic vein or portal vein thrombosis.19,20
This present study also analysed spectral Doppler waveforms in the hepatic veins after liver transplantation. Hepatic venous waveforms are normally triphasic, with two antegrade waves related to atrial and ventricular diastole, and one retrograde wave caused by the increase in pressure in the right atrium during atrial systole. Absence of this retrograde flow has been associated with venous stenosis and acute graft rejection.21–23 All three blood flow patterns (triphasic, biphasic and monophasic waveforms) were detected in the hepatic veins of liver recipients after transplantation without complications in the present study. The proportions of monophasic waveforms found in the right, middle and left hepatic veins were 32.6%, 34.7% and 22.2%, respectively. So the abnormalities in the hepatic vein waveforms were generally not specific to a diagnosis of hepatic vein stenosis. Graft oedema and changes in cardiac output may also contribute to abnormal Doppler waveforms in the hepatic vein.24,25
In summary, as a result of the present findings of the follow-up of the haemodynamic changes in the normally transplanted liver, we recommend that the appropriate follow-up time-points after liver transplantation with ultrasonography are: once a day during the first 7 days, once a week for the following 2–4 weeks, then Once every 3 months for up to 1 year, and then once every 6 months. During follow-up with ultrasonography, a high HARI arterial flow visualized in the transplanted liver does not mean the occurrence of hepatic arterial complications. Observation of a low HARI intrahepatic arterial flow, however, should be considered indicative of anastomotic stenosis of the common hepatic artery and further digital subtraction angiography examination is recommended. During the follow-up examination with ultrasound, decreased portal vein blood flow in the transplanted liver is a normal haemodynamic change in the first 6 months after liver transplantation. However, if the velocity of the portal vein is not decreased, then this might indicate the possibility of hepatic arterial complications in the transplanted liver because of the close relationship between the portal vein and hepatic artery. If this situation arises, a more detailed follow-up examination with colour Doppler ultrasonography is recommended.
In conclusion, this study documented the range of the changes observed in the haemodynamic data measured in the transplanted liver without complications. It also provided information about hepatic haemodynamic changes at different time-points after liver transplantation. The typical hepatic haemodynamic changes that were observed were (i) a markedly decreased portal vein peak velocity within 1 month after liver transplantation; and (ii) a more gradual decrease in hepatic artery peak systolic velocity over the 3 years post-transplantation. The most striking findings were that such abnormal Doppler results and blood flow waveforms in liver recipients after transplantation were not always indicative of complications. Abnormal Doppler results at follow-up should be interpreted with caution, because they may not be clinically significant and may gradually and spontaneously improve.
Footnotes
Declaration of conflicting interest
The authors declare that there are no conflicts of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.