Page 17 - Read Online

Pais-Costa et al. Mini-invasive Surg 2018;2:33 I http://dx.doi.org/10.20517/2574-1225.2018.33 Page 5 of 11

Table 1. Patient Characteristics
Variable Value
Gender, female, n (%) 21 (70)
Age, years, median (range) 38 (21-71)
ASA risk score, n (%)
1 28 (93)
2 2 (7)
3
Previous open abdominal surgery, n (%) 6 (20)
Etiology, n (%)
Adenoma 16 (50)
Cystadenoma 5 (17)
Focal nodular hyperplasia 4 (13)
Pyogenic liver abscess 3 (10)
Hemangioma 3 (10)
Non-oriental hepatolithiasis 1 (3)
Number of lesions, median (range) 1 (1-6)
Lesions size, cm, median (range) 5 (3-13)

[16]
was defined in accordance with the Clavien-Dindo classification of surgical complications . Biliary fistula
[17]
was defined using the International Study Group of Liver Surgery classification . Follow-up included
physical examinations and hepatic blood tests every three months, as well as abdominal imaging (CT or
MRI) twice a year. Recurrence was considered when a lesion reappeared on imaging examinations and was
finally confirmed by means of histological analysis. QoL was measured by means of a QoL questionnaire
[15]
using the Short-Form Health Survey (SF 36), as previously reported by Giuliani et al. . This test was
administered by means of a personal interview before surgery and at 1, 3, 6 and 12 months after surgery. In
addition, a simple non-validated questionnaire was also performed including the following questions: (1)
satisfaction of surgical scar; (2) exposure of body image when wearing bikinis, shorts, “summer clothes”,
etc.; (3) change of daily behavior (social, type of the clothes, sportive practices); and (4) recommendation of
the LH to other patients. The responses were classified into positive or negative. All these questions were
considered to evaluate body image satisfaction regarding the esthetics aspects after LH.

Statistical Analyses
Categorical variables are described using percentages. Continuous variables are expressed as means and
standard deviation for symmetrically distributed, and median (range) for nonsymmetrically distributed
data. Differences in SF-36 QoL questionnaire scores between preoperative evaluations and postoperative
evaluations were analyzed using two-way repeated measures ANOVA and each patient was in his or her
own control at the time of the evaluation. There were no significant outliers in pairwise comparisons and
normality assumption held at Shapiro-Wilk test using residuals. We used Wilks’ lambda as an omnibus
test statistics for differences between the means of QoL evaluations. Within-subjects effects were evaluated
by F-test with Greenhouse-Geisser correction due to unequal variances of the differences between all
combinations of related groups, as assessed by Mauchly’s test. Post hoc pairwise comparisons were
calculated with Bonferroni adjustment for multiple comparisons. We reported exact P values up to 0.001
and the significance level was set to 0.05 (5%) for statistical inference. Statistical analyses were performed
using PASW Statistics 18 (SPSS Inc., Hong Kong, 2015).

RESULTS
The general characteristics of the study population are detailed in Table 1. Median age was 38 years (range
21-71) and the majority were females (70%). The most frequent etiology was HA in 16 patients (50%).
Among symptomatic patients (n = 3) who underwent LH, two of them had giant hepatic hemangiomas
while the third patient had a giant focal nodular hyperplasia (FNH). Their symptoms were as follows: