Retrospective analysis based on a clinical grading system for patients with hepatic hemangioma: A single center experience

Abstract

BACKGROUND

The optimal approach for managing hepatic hemangioma is controversial.

AIM

To evaluate a clinical grading system for management of hepatic hemangioma based on our 17-year of single institution experience.

METHODS

A clinical grading system was retrospectively applied to 1171 patients with hepatic hemangioma from January 2002 to December 2018. Patients were classified into four groups based on the clinical grading system and treatment: (1) Observation group with score < 4 (Obs score < 4); (2) Surgical group with score < 4 (Sur score < 4); (3) Observation group with score ≥ 4 (Obs score ≥ 4); and (4) Surgical group with score ≥ 4 (Sur score ≥ 4). The clinico-pathological index and outcomes were evaluated.

RESULTS

There were significantly fewer symptomatic patients in surgical groups (Sur score ≥ 4 vs Obs score ≥ 4, P < 0.001; Sur score < 4 vs Obs score < 4, χ² = 8.60, P = 0.004; Sur score ≥ 4 vs Obs score < 4, P < 0.001). The patients in Sur score ≥ 4 had a lower rate of in need for intervention and total patients with adverse event than in Obs score ≥ 4 (P < 0.001; P < 0.001). Nevertheless, there was no significant difference in need for intervention and total patients with adverse event between the Sur score < 4 and Obs score < 4 (P > 0.05; χ² = 1.68, P > 0.05).

CONCLUSION

This clinical grading system appeared as a practical tool for hepatic hemangioma. Surgery can be suggested for patients with a score ≥ 4. For those with < 4, follow-up should be proposed.

Key Words: Hepatic hemangioma; Clinical grading system; Surgical indication; Outcome; Postoperative complications

Core Tip: With the development of surgical technique and new intervention including transcatheter arterial embolization and radiofrequency ablation used in liver surgery, the incidence of postoperative complications has been significantly decreased, however, the treatment of hepatic hemangioma still needs to consider the balance of benefit and risk, surgical indications for hepatic hemangioma remain unclear. Here, we evaluate a clinical grading system for management of hepatic hemangioma based on our 17 years of experience. The clinical grading system combined with the individual situation of patients could be helpful to select the most appropriate treatment for these lesions.

INTRODUCTION

Hepatic hemangiomas are the most common benign tumors in the liver, accounting for approximately 73% of all cases, with an incidence of 0.4%-7.3% at autopsy and an incidence of 1.7% by abdominal ultrasound (US) examination[1-4]. Although most hemangiomas are small and stable in the follow-up, a small subset of lesions may cause abdominal discomfort and life-threatening complications, such as Kasabach-Merritt syndrome, rupture and internal hemorrhage[5-11]. A minority of patient were advised to receive liver transplantation due to giant hemangiomas causing severe symptoms and cannot be treated otherwise[12-14]. While surgery is the most effective and radical intervention for hemangioma, prophylactic hepatectomy is not recommended to prevent complications. However, owing to the lack of high-quality evidence-based consensus or guidelines, the optimal approach for managing hepatic hemangioma remains controversial.

The natural history and risk of complications of hepatic hemangioma are unknown, and surgery is advocated for most hepatic hemangiomas owing to concerns regarding possible rupture, large tumor size, anxiety, and lack of valid criteria for treatment in China. To clarify the proper surgical indications, we gathered experts among surgeons, interventional radiologists, imaging, pathologists, and case managers as a multidisciplinary team (MDT) to provide a clinical grading system for patients with hepatic hemangioma in 2015 (Table 1)[15]. The objective of the present large retrospective study was to evaluate the proposed clinical grading system through our single-center experience in the management of 1171 patients with hepatic hemangioma.

Table 1 Determination of hepatic hemangioma grade according to the clinical grading system.

Points assigned	0	1	2
Symptom	No	Mild	Obvious
Diameter (cm)	< 5	5-10	≥ 10
Location, segment	II, III, VI, VII	IV, V, VIII	I
Tumor increment (cm/year)	0	< 1	≥ 1

MATERIALS AND METHODS

Proposal of a clinical grading system for hepatic hemangioma

The clinical grading system for hepatic hemangioma was proposed by the MDT based on the clinical effects concerning symptoms, diameter, location, and tumor increment (Table 1)[15]. Surgery was proposed for patients with a score ≥ 4, otherwise, observation, transcatheter arterial embolization (TAE), or radiofrequency ablation (RFA) was considered alternative choices for inoperable patients; observation was proposed for patients with a score < 4. Besides, surgical therapy should be conducted given the risk of severe and fatal complications of hepatic hemangioma, such as Kasabach-Merritt syndrome, rupture, Budd-Chiari syndrome, jaundice, and heart failure.

Patient subgrouping and application of the clinical grading system

Based upon abovementioned grading system and treatment method, a total of 1171 patients with hepatic hemangioma who were assessed in the Digestive and Vascular Surgery Center, the First Affiliated Hospital of Xinjiang Medical University between January 1, 2002 and December 31, 2018 were categorized into four groups: (1) Observation group with score < 4 (Obs score < 4, n = 352); (2) Surgical group with score < 4 (Sur score < 4, n = 460); (3) Observation group with score ≥ 4 (Obs score ≥ 4, n = 77); and (4) Surgical group with score ≥ 4 (Sur score ≥ 4, n = 282). The patients’ medical records were reviewed to collect the demographic data, laboratory values, imaging studies, tumor characteristics, mode of operation, clinical and/or postoperative outcomes, and complications. The patients’ characteristics are shown in Table 2.

Table 2 Clinical characteristics and laboratory data of the patients with hepatic hemangioma.

Characteristics	Score < 4		Score ≥ 4		F/χ2	P value
	Observation (n = 352)	Surgery (n = 460)	Observation (n = 77)	Surgery (n = 282)
Age (year)	46 (39, 52)	44 (39, 51)	42 (36, 54)	45 (39, 50)	0.55	0.65
HBV positive, n (%)	18 (5.11)	34 (7.39)	6 (7.79)	17 (6.03)	-	0.54
HCV positive, n (%)	3 (0.85)	1 (0.22)	0	1 (0.35)	-	0.61
Total bilirubin (μmol/L)	11.80 (9.25, 15.71)	11.76 (9.17, 15.19)	12.70 (9.42, 16.27)	11.50 (9.41, 11.51)	0.92	0.43
Albumin (g/L)	40.09 (37.63, 43)	40.95 (38, 43.9)	40 (37.65, 46.88)	40.51 (37.7, 43.51)	1.04	0.37
AST (U/L)	18.70 (15.6, 23.95)	18.20 (15.43, 22)	18 (15.3, 22.55)	18.5 (15.78, 21.9)	0.89	0.45
ALT (U/L)	17.7 (13, 25.2)	17.55 (12.85, 24.42)	16.2 (11.65, 23.2)	17.6 (12.7, 24.7)	0.46	0.71
Child-Pugh Grade A, n (%)	349 (99.15)	452 (98.26)	75 (97.40)	278 (98.58)	-	0.51
Coagulopathy (n)	0	0	0	0	-	-

HBV: Hepatitis B virus; HCV: Hepatitis C virus; AST: Aspartate aminotransferase; ALT: Alanine aminotransferase.

Follow-up method

Follow-up was carried out via outpatient examination (including physical examination, liver function tests and abdominal US) and telephone interviews annually thereafter. computed tomography or magnetic resonance imaging can be performed if necessary. Patients were followed through December 2022 with a mean follow up time of 40 months (range from 24 to 208 months).

Definitions

Postoperative complications were recorded to 30 days after operation. The criteria for postoperative bleeding[16], bile leakage[17], and other perioperative complications were defined by the Clavien-Dindo classification[18].

Statistical analysis

Statistical analysis was performed using SPSS statistical software (Version 23.0, SPSS, Chicago, IL, United States). Continuous variables are expressed as the mean ± SD or median with interquartile range, and differences were analyzed with the Kruskal-Wallis test. Categorical variables were analyzed with χ² test or Fisher’s exact test. A P value < 0.05 was considered statistically significant.

RESULTS

Patient characteristics

The study population comprised 739 women (63.1%) and 432 men (36.9%), with a median age of 45 years (range, 20-82 years). 626 (53.5%) patients had solitary hemangioma, and 545 (46.5%) cases had at least two lesions. 445 (38.0%) and 702 (60.0%) patients presented a stable lesion size with no enlargement and a slow increase (< 1 cm/year) respectively during follow-up, only 24 (2%) presented with rapidly growing lesions (≥ 1 cm/year). The baseline characteristics of four groups are shown in Table 2. No significant difference were found in age, incidence of hepatitis B virus/hepatitis C virus positive, and liver function between the four groups.

Comparison of outcomes between Obs score ≥ 4 and Sur score ≥ 4

In Sur score ≥ 4, 274 (97.2%) patients were free from abdominal discomfort postoperatively, while only eight patients (2.8%) had persistent symptoms after surgery. Postoperative complications occurred in 26 (9.2%) cases, and included biliary leakage in six patients, hydrothorax in 13 patients, abdominal abscess in one patient, pulmonary infection and urinary infection in one patient each, and postoperative abdominal bleeding in four patients. Reoperation were conducted in two patients due to postoperative abdominal bleeding, and another two patients died as a result of massive hemorrhage (Table 3). The postoperative complications was classified according to the Clavien-Dindo system (Table 4). Residual or recurrence of hemangioma occurred in three patients after surgery for whom TAE was performed. In addition, one patient with incisional hernia was referred after 12 months by reconstruction with mesh. On the other hand, in Obs score ≥ 4, 77 patients were treated by clinical observation, 34 (44.2%) patients had no abdominal complaints, and continuous or intensified or new onset of symptoms occurred in 43 (55.8%) patients. During follow-up, intervention was necessary in 31 patients, including surgery in two patients, TAE in 27 patients and RFA in two patients. No patient died related to hepatic hemangioma. Compared with Obs score ≥ 4, there were significantly fewer symptomatic patients in Sur score ≥ 4 (P < 0.001) (Table 3).

Table 3 Comparison of outcomes during follow-up between the surgical and observation group classified by score ≥ 4 and score < 4, n (%).

	Score < 4 (n = 812)				Score ≥ 4 (n = 359)
	Observation (n = 352)	Surgery (n = 460)	χ2	P value	Observation (n = 77)	Surgery (n = 282)	χ2	P value
No abdominal complaints	308 (87.5)	430 (93.5)	8.60	0.004	34 (44.2)	274 (97.2)	-	< 0.001
Continuous or intensified or new onset of abdominal complaints	44 (12.5)	30 (6.5)			43 (55.8)	8 (2.8)
Postoperative complications	-	54 (11.7)			-	26 (9.2)
Hepatic	-	12 (2.6)			-	11 (3.9)
Extrahepatic	-	42 (9.1)			-	15 (5.3)
Need for intervention	15 (4.3)	8 (1.7)	-	0.052	31 (40.3)	5 (1.8)	-	< 0.001
Operation	1 (0.3)	1 (0.2)			2 (2.6)	2 (0.7)
TAE	5 (1.4)	2 (0.4)			27 (35.1)	3 (1.1)
RFA	9 (2.6)	5 (1.1)			2 (2.6)	0 (0)
Mortality related to hepatic hemangioma	0 (0)	0 (0)			0 (0)	2 (0.7)		-
Total patients with adverse event	48 (13.6)	78 (17.0)	1.68	0.20	45 (58.4)	34 (12.1)	-	< 0.001

More than one type of adverse outcome per patient is possible. TAE: Transcatheter arterial embolization; RFA: Radiofrequency ablation.

Table 4 Postoperative complications according to the Clavien-Dindo classification, n (%).

	Score < 4 (n = 460)	Score ≥ 4 (n = 282)
Grade I	30 (6.5)	6 (2.1)
Grade II	1 (0.2)	2 (0.7)
Grade IIIa	21 (4.6)	14 (5.0)
Grade IIIb	1 (0.2)	2 (0.7)
Grade IV	0 (0)	0 (0)
Grade V	0 (0)	2 (0.7)

The adverse events are described in detail in Table 3. For the observation group, the adverse events included continuous or intensified or new onset of abdominal complaints, need for intervention and mortality related to hepatic hemangioma. For the surgical group, the adverse events included postoperative complications, and other events as described in observation group. Despite a higher rate of postoperative complications initially, the patients in Sur score ≥ 4 had a lower rate in need for intervention and total patients with adverse event than in Obs score ≥ 4 (P < 0.001; P < 0.001).

Comparison of outcomes between Obs score < 4 and Sur score < 4

In Sur score < 4, 430 (93.5%) patients were free from abdominal discomfort postoperatively, 28 (6.1%) patients had persistent symptoms after surgery, and two (0.4%) patients had a new onset of abdominal pain after surgery (Table 3). Postoperative complications occurred in 54 (11.7%) cases, including postoperative abdominal infection and adhesive intestinal obstruction requiring reoperation in one patient, biliary leakage in 11 patients, hydrothorax in 21 patients, wound infection in eight patients, incisional fat liquefaction in 11 patients, and pulmonary infection in one patient. Residual or recurrence of hemangioma occurred in seven patients after operation, of which two and five cases underwent TAE and RFA, respectively. No perioperative deaths occurred. In addition, three patients with incisional hernia were referred after the operation (range, 7-18 months) and all cases were treated by reconstruction with mesh. On the other hand, in Obs score < 4, 308 (87.5%) patients had no abdominal complaints, and 44 (12.5%) patients had continuous or intensified or new onset of abdominal symptoms. 15 patients needed for intervention, of which one, five and nine patients accepted surgery, TAE and RFA respectively. No patient died related to hepatic hemangioma. Compared with Obs score < 4, there were significantly fewer symptomatic patients in Sur score < 4 (χ² = 8.60, P = 0.004). Meanwhile, there was no statistically significant difference in need for intervention and total patients with adverse event between Obs score < 4 and Sur score < 4 (P > 0.05; χ² = 1.68, P > 0.05) (Table 3).

Comparison of outcomes between Obs score < 4 and Sur score ≥ 4

There were significantly fewer symptomatic patients in Sur score ≥ 4 than in Obs score < 4 (P < 0.001). Nevertheless, there was no significant difference in need for intervention and total patients with adverse event between the two groups (P > 0.05) (Table 5).

Table 5 Comparison of outcomes between the observation score < 4 and surgery score ≥ 4, n (%).

	No abdominal complaints	Continuous or intensified or new onset of abdominal complaints	Postoperative complications	Need for intervention	Mortality related to hepatic hemangioma	Total patients with adverse event
Score < 4 observation (n = 352)	308 (87.5)	44 (12.5)	-	15 (4.3)	0	48 (13.6)
Score ≥ 4 surgery (n = 282)	274 (97.2)	8 (2.8)	26 (9.2)	5 (1.8)	2 (0.7)	34 (12.1)
P value	< 0.001	< 0.001	-	0.060	-	0.634

DISCUSSION

To the best of our knowledge, this is the first report regarding the largest scale patients with 1171 hepatic hemangioma assessing the clinical outcomes based upon previously proposed grading system[15]. The current data advocate surgical resection for tumor score ≥ 4, while observation is recommend when score < 4.

Surgical indications for hepatic hemangioma remain unclear, and the optimal approach for managing hepatic hemangioma is controversial. Symptoms, tumor diameter, location, and tumor increment are of the greatest significance for treatment of hepatic hemangioma, and thus these variables comprise the proposed clinical grading system. The implementation of this system could not only avoid prophylactic hepatectomy, but also provide the threshold of surgical indication. The proposed grading system is simple in design, easy to operate, and practical for devising the most appropriate strategy for hepatic hemangioma, it has strong clinical value.

Kasabach-Merritt syndrome and rupture are indications for surgery, and the mortality rate was up to 30% and 60%, respectively[2,19]. Some patients with giant hemangiomas were advised to receive liver transplantation as a unique option[11-13]. Meanwhile, the disadvantages of observation was found to be associated with 56% of new onset symptoms, 9% of major complications, and 1.4% of deaths owing to tumor rupture[20]. So excessive emphasis on observation will delay the best time for resection. However, prophylactic hepatectomy is not recommended to prevent complications for hemangioma. The proposed grading system based on MDT discussion and long-term clinical practice was developed in 2015, it is believed that a score of 4 points is a critical value, with indications for surgery if it is greater than or equal to 4 points, and observation could be recommended if it was less than 4 points. In this large retrospective study, the incidence of abdominal complaints, postoperative complications, need for intervention, and total patients with adverse event in Sur score ≥ 4 was 2.8%, 9.2%, 1.8%, and 12.1% respectively. In contrast, the Obs score ≥ 4 was associated with 55.8% of continuous or intensified or new onset of abdominal symptoms, 40.3% of need for intervention, and 58.4% of total patients with adverse event. The findings of this study demonstrate that patients with score ≥ 4 may achieve greater benefits from surgery (P < 0.001). On the other hand, there was no significant difference in need for intervention and total patients with adverse event between Sur score < 4 and Obs score < 4 ( P = 0.205). This study effectively confirm that the proposed grading system has a potential clinical significance.

In this present study, two patients during our preliminary experience stage with giant hemangioma lesions in both of liver lobe had experienced extended hepatic resection. The two patients with score = 6 died of unexpected massive haemorrhage and post-operative coagulation dysfunction due to a lack of complete and systematic preoperative evaluation. Thanks to the development of surgical technique including staged-hepatectomy, low central venous pressure control and meticulous bleeding control, there are no post-operative death experienced even in patients with giant hemangioma in our later experience since 2012. In addition, another two patients with scores = 7 had no chance for surgery due to giant tumor size (≥ 25 cm) and complicated location, and liver transplantation was the only option. Therefore, surgery was recommended for hepatic hemangioma with a score ≥ 4, excessive emphasis on observation will delay the best time for resection. For those with a score ≥ 6, hepatectomy should be performed in high-quality centers by experienced hepatic surgeons to avoid serious morbidity and mortality. If more than one giant tumors exists, staged hepatectomy can be adopted according to the intraoperative situation.

In the current study, the data of 1171 patients with hepatic hemangioma were analyzed with the clinical grading system. The number of enrolled patients (n = 1171) was significantly higher than the highest reported comparative study previously (n = 556)[21]. The retrospective study design and single-center non-randomized study are the major drawbacks of this study. Nevertheless, a single-center study has the advantage of forming a preliminary surgical indication and consistency of surgical procedures.

CONCLUSION

The proposed clinical grading system is practical and effective. The score 4 as a threshold, combined with the individual situation of patients could be helpful to select the most appropriate treatment for hepatic hemangioma. Surgery can be considered the most effective and radical intervention for hepatic hemangioma with a score ≥ 4. In cases with a score < 4, follow-up should be proposed. The clinical grading system appears to represent a practical tool for devising a better strategy for hepatic hemangioma, although its value requires further evaluation in multi-center studies.

ACKNOWLEDGEMENTS

We thank all the patients who participated in this study and all the medical workers who helped us in this work. The relating text has been added to the manuscript.