The portal vein is responsible for close to two thirds of the hepatic blood flow. It supplies blood rich in oxygen, nutrients, growth factors, and hormones, among other elements. The portal vein only resists low pressures and portal hypertension is defined when portal pressure exceeds 5mmHg. A continuous flow with a normal range between 15-30cm/s can be observed through Doppler ultrasound.1,2

The etiology of portal hypertension syndrome is varied. Based on its anatomic classification, it can be categorized into prehepatic, intrahepatic (presinusoidal, sinusoidal, and postsinusoidal), and posthepatic. The reported prevalence of each etiology is varied. In their studies, Poddar et al.3 and Donatone4 reported a predominance in frequency of prehepatic portal hypertension over hepatic portal hypertension: 68% versus 32% and 54% versus 46%, respectively. Disease manifestations of prehepatic portal hypertension have been described to present at an initial age of 4.6 years and those of hepatic portal hypertension at 6.9 years.

The main causes of prehepatic portal hypertension are: portal vein thrombosis, cavernomatous degeneration of the portal vein, and splenic vein thrombosis. Other causes, to a lesser degree, are congenital malformations of the portal vein and Klatskin tumor. The first 2 of the abovementioned pathologies are the most studied. The primary initial clinical manifestation of prehepatic portal hypertension is upper gastrointestinal bleeding; close to 90-95% of these patients have esophageal varices and 35-40% have gastric varices.5

Doppler USG, splenoportography, magnetic resonance angiography, and angiotomography are among the diagnostic studies utilized for this disease.

There are 2 important treatment aspects: a)medical treatment, with medications like propranolol, ranitidine, or a proton pump inhibitor; 2)endoscopic treatment of varices; there are few pediatric studies comparing ligation versus sclerotherapy, and 3)surgical treatment, with portosystemic shunts (total or selective), splenectomy, or partial embolization.

The primary aim of this study was to determine the main clinical, radiologic, and endoscopic characteristics of patients upon diagnosis with prehepatic portal hypertension at the Instituto Nacional de Pediatría within the time frame of January 2001 and December 2011.

An observational, descriptive, retrolective, and cross-sectional study was conducted. The protocol was accepted by the Research and Ethics Committee of the Instituto Nacional de Pediatría. A search of the Clinical Archive database of all the medical records with confirmed diagnosis was carried out. Fifty-two clinical case records were obtained of patients seen at the institution from January 1, 2001 to December 31, 2011. Of these, 32 had prehepatic etiology and were included in the review. Demographic, clinical, imaging, and treatment (medical and surgical) variables were analyzed. Employing the SPSS software version 19.0, the quantitative variables were analyzed through the descriptive statistics of measures of central tendency, dispersion, and normality tests and the categorical variables through frequencies and percentages.

The diagnoses contained in the 52 medical records initially reviewed were distributed as follows:

Prehepatic portal hypertension: 32 patients (61.5%).

Presinusoidal intrahepatic portal hypertension: 6 patients (11.5%), all diagnosed with hepatoportal sclerosis.

Sinusoidal intrahepatic portal hypertension: 12 patients (23%). The reported diagnoses were: biliary atresia (5), autoimmune hepatitis (3), neonatal hepatitis (1), lysosomal disease (1), liver abscess (1), and alpha1 antitrypsin deficiency (1).

Posthepatic portal hypertension: 2 patients (4%): one patient with scimitar syndrome and the other with altered flow of the inferior vena cava.

Thirty-two patients diagnosed with prehepatic portal hypertension were included in this study: 12 patients (37.5%) were female and 20 (62.5%) were male. The median age at the time of diagnosis was 107 months (minimum of 3 months and maximum of 181 months), of which 4 children (12.5%) were 0 to 12 months old, 15 (46.88%) between 1 and 4 years old, 7 (21.88%) between 5 and 9 years old, 5 (15.62%) between 10 and 13 years old, and one case (3.12%) between 14 and 18 years old. The median weight was 17kg (a minimum of 4.06 and maximum of 62kg). The median height was 102cm (a minimum of 58cm and maximum of 162cm).

In our study, we found a 61.5% prevalence of prehepatic etiology in the portal hypertension cases, which is in accordance with the values described in the literature.3,4 Nevertheless, we believe there is an under-registering of cases of biliary atresia and other etiologies that progress to cirrhosis, and consequently to cases of intrahepatic portal hypertension; therefore we cannot affirm that prehepatic portal hypertension is the most frequent cause of all the portal hypertensions.

We found a higher frequency in males (62.5%) in our case series, which was interesting, given that this predisposition is not observed in other published reports. At the time of diagnosis, the values of the age, weight, and height variables were very different, which also is in contrast with that reported in the literature; in the majority of the studies, the patients are older than 4years of age, whereas in our series, 19 patients (59%) were diagnosed under the age of 4years.3–9

Upper gastrointestinal bleeding secondary to esophageal varices was the primary clinical manifestation at the time of diagnosis, correlating with that reported in the literature.3–9 Bleeding together with anemia was found in the majority of the cases (71.9%) and bleeding without anemia in 6.3%. This also correlates with studies in the literature stating that at the time of diagnosis, the majority of patients require transfusions of blood derivatives in order to achieve hemodynamic stability, resulting in a 2 to 5% mortality rate.5 It is striking that in the patients that were initially evaluated due to hypersplenism or isolated splenomegaly, the etiology was regarded as hematologic and oncologic. However, it should be kept in mind that that the cause could be prehepatic portal hypertension, as occurred with our patients. The leading case series7,8,10 describe cavernomatous degeneration and portal vein thrombosis as the main etiologic factors of this pathology, which is in accordance with the findings of our case series, in which we found that 84.4% of the cases, as a whole, corresponded to these pathologies.

Doppler ultrasound is the first imaging study that should be carried out in the initial approach to these patients,11 because it enables the measuring of flow velocity (measured in cm/s); according to the literature, the diagnosis of portal hypertension is made when there are values above 25-30. In our study, flow velocity was measured in just 13 of the patients, only 3 of which met the portal hypertension criterion of that value described in the literature. This is also an interesting finding, given that high sensitivity and specificity have been reported for this diagnostic study. However, it is possible that these values were actually higher, if we take into account the fact that the majority of patients presented with gradeii and iii esophageal varices at the time of diagnosis. Published reports indicate that USG reliably shows the presence of collateral veins and enables a nonspecific determination of the flow obstruction site. This concurs with our findings of collateral vessels (84.4% of the imaging studies) and blood flow obstruction (56.3%).5,11

Splenoportography was performed on approximately half of the documented patients (53%), and was the main diagnostic study in this group of patients from 2001 to 2006, the first 5years of the time frame of our analysis. As described in the literature, this technique stopped being used and has been replaced by angiotomography and magnetic resonance angiography because they provide the same information and are noninvasive studies,11–13 explaining why splenoportography was not indicated for the patients at the Instituto Nacional de Pediatría during the period of time corresponding to the last 5years of our study time frame.

There are no studies in the literature that compare angiotomography and magnetic resonance angiography and indicate that one diagnostic method is better than the other for this pathology. Both methods define images characteristic of collateral vessels, flow obstruction, spontaneous shunts, and splenomegaly; magnetic resonance angiography has a slightly higher sensitivity and specificity,11 depending on the researcher and the institution carrying out the study. Magnetic resonance angiography is performed more frequently (43.75%) than angiotomography; collateral vessels and splenomegaly were the most frequent findings in each of these methods.

Propranolol is mentioned in the literature as a regular part of medical treatment,14–17 and was indicated for 100% of the study patients. In recent studies, proton pump inhibitors are used to inhibit hydrochloric acid (taking into account their mechanism of action). Nevertheless, in our institution they were only prescribed to 65.5% of the patients. Ranitidine was indicated for the rest of the patients; this drug has not been the treatment of choice for the last few years, not only because of its limited action compared to that of the proton pump inhibitors, but also due to its high risk for tachyphylaxis.

Upper gastrointestinal bleeding secondary to esophageal varices is regularly described in the literature, reporting a prevalence of up to 90-95% in the patients with prehepatic portal hypertension.5 In our study, the presence of gradeii and iii esophageal varices made up 97% of the cases, explaining the high risk for bleeding and the reason for the first medical consultation. It has been stated that gastric varices may be present in up to 40% of cases; gastroesophageal varices were found in 37.5% of cases at the Instituto Nacional de Pediatría, but not a single case of isolated gastric varices. Gastropathy has been directly correlated with the degree of liver involvement;5 almost all of these patients present with gradeii and iii esophageal varices, which is directly correlated with the 75% prevalence of congestive gastropathy in this study.

The use of ligatures or sclerotherapy in the treatment of esophageal varices is controversial and there are few pediatric studies on this subject. Some reports have described both procedures as having long-term management efficacy, but ligation has been shown to reduce the number of sessions required and has a lower percentage of bleeding recurrence and secondary complications.16–18 In the present study, ligation was documented in 8 patients and sclerotherapy in 5. In the follow-up period, 4 more patients received sclerotherapy treatment. This is explained by the fact that in the first 5years covered in our review (2001 to 2006) the patients were quickly treated with portosystemic shunt placement, considerably reducing the grade of esophageal varices, and thus resulting in a smaller number of procedures. It is interesting to point out that although ligatures have been available at the Institute for only 5years, this procedure has been carried out more frequently than sclerotherapy, according to the data we collected.

We found no study indicating a predominance of a specific portosystemic shunt in this disease. They are recommended to always be evaluated individually, after documenting the collateral vessels, their distribution, and flow velocities and they should not be considered an emergency procedure.15,19 Portosystemic shunt was carried out on 17 cases (53%) in the present analysis. It is striking that all of the patients receiving a mesocaval shunt underwent this procedure during the first month of evaluation. Certain advantages to distal splenorenal shunt have been reported in relation to the rest of the total shunts,6,19 but mesocaval shunt (10 cases) predominated over distal splenorenal shunt (7 cases) in this study. The surgical option of splenectomy was contemplated in 9 patients, though it is not a first choice procedure in the follow-up of these patients.6

Prehepatic portal hypertension is an entity with a high morbidity rate, and therefore it is important that the physician be familiarized with this disease so that the patient can be referred to a tertiary care center for opportune diagnosis and treatment.

No financial support was received in relation to this study/article.

The authors declare that there is no conflict of interest.