Page 290 - Read Online

Butt et al. TAE for ruptured HCC in Pakistan

Modality of ruptured HCC diagnosis was defined reviewed. A total of 24 patients were diagnosed to
as: (1) incidental when an asymptomatic HCC was have spontaneously ruptured, unresectable HCC
discovered on imaging done during diagnostic and were analyzed. The mean age was 58.29 ±
procedures performed for some other disease; or (2) 15.25 years (range 17-93 years) and most of them
symptomatic when diagnosed during workup after 21 (87.5%) were males. Hepatitis C was the most
symptom appearance. The HCC was considered as common cause of cirrhosis (79.2% cases). The mean
“non-advanced” if the lesion was solitary ≤ 5 cm or Child-Pugh score was 9.96 ± 2.85 (range 7-15) and
paucifocal ≤ 3 lesions, with the largest diameter ≤ mean MELD score was 17.92 ± 6.38 (range 9-32).
3 cm, in the absence of vascular invasion and distant On presentation 62.5% had decompensated cirrhosis
metastases or “advanced,” when the tumor exceeded and many of them had prior history of hospitalization
these limits. Moreover, the HCC was also classified with spontaneous bacterial peritonitis (16.7%),
for macroscopic types as: (1) solitary; (2) paucifocal (≤ portosystemic encephalopathy (20.8%), variceal
3 nodules); (3) multifocal (> 3 nodules); (4) infiltrative bleed (12.5%) or hepatorenal syndrome (4.2%). The
(infiltrating pattern of HCC); or (5) massive (huge most common clinical manifestations of ruptured HCC
mass with a diameter of > 10 cm and an undefined on presentation were sudden abdominal pain (83.3%),
boundaries). [17] In the presence of ≥ 2 lesions, the hemoperitoneum (54.2%), symptoms of anemia
largest tumor was considered as representative of (83.3%) and hypovolemic shock (25.0%). Diagnosis
HCC and the diameter of the representative tumor of ruptured HCC was confirmed on CT scan of
measured in its greatest dimension was recorded as abdomen in all cases. The mean tumor size was 7.76
tumor size. Furthermore, information was recorded ± 4.22 cm (range 1.7-17.7 cm). Almost two-third of
regarding hepatic lobes involved, presence of portal patients had multifocal (50.0%) or massive/infiltrative
vein thrombosis and extra hepatic spread. (25.0%) HCC. Moreover, advanced HCC was found
in 87.5% cases on presentation [Table 1].
The patients were treated “conservatively” when liver
reserves were poor defined by a Child class C or they A total of 11 (45.8%) patients were treated conservatively
were severely ill due to other comorbid conditions. who either had poor general condition, impaired hepatic
TAE was performed in a well-equipped interventional reserves, multiple lesions, or when patient had declined
radiological suite by a team of experienced any intervention. TAE was performed in 13 (54.8%)
interventional radiologists and Gel foam was used cases of ruptured HCC. None of them underwent for
as embolizing agent. The study was conducted by emergency resection.
maintaining compliance with the Helsinki Declaration
and was approved by the Ethical review committee of Comparison of patients treated
Aga Khan University Hospital, Karachi. conservatively vs. those treated with TAE
There was no statistically significant difference in age,
Statistical analysis gender, etiology of underlying cirrhosis or symptoms
Data was entered and analyzed in SPSS version and signs at presentation among those treated
17.0. Mean ± SD and ranges were calculated for conservatively as compared to those who underwent
continuous variables and proportions for categorical TAE. The tumor size, macroscopic types, location
variables. To see the difference between two groups and stage of HCC were also comparable among
independent student t-test, Chi square or Fisher exact both groups [Table 2]. Although the prior hepatic
was used where appropriate. A univariate logistic decompensations, MELD and Child score were
regression analysis was conducted to assess the comparable in both groups, most of the patients in
(crude) association of the prognostic factors for 30- conservative group had patients with Child class C as
day mortality. Biological significance and a value of P compared to TAE group (54.5% vs. 15.4%, P = 0.08).
≤ 0.1 were considered as criteria for a variable to be Likewise, serum total bilirubin level (5.14 ± 3.50 vs. 2.15
significant at univariate analysis. Biological plausible
interactions among variables and confounding were ± 1.04, P = 0.008) was higher and albumin was lower
also checked. Multivariable logistic regression was (2.04 ± 0.41 vs. 2.63 ± 0.49, P = 0.004) in conservative
done and results are expressed as odds ratio (OR), treatment group as compared to TAE group.
along with 95% confidence interval (CI).
The control of HCC bleeding was achieved in 66.7%
RESULTS cases which was significantly higher for those who
were treated via TAE as compared to those who
Clinical characteristics of patients were treated conservatively (92.3% vs. 36.4%, P =
The medical records of 850 patients with HCC who 0.008). Overall median duration for which the patients
had visited our center during the study period were remained alive after HCC rupture was longer for
Hepatoma Research ¦ Volume 2 ¦ October 21, 2016 281