Received: 28 December 2021 | Accepted: 24 February 2022 DOI: 10.1002/lt.26452 BRIEF REPORT Sinusoidal obstruction syndrome after liver transplantation: A multicenter observational study To the editor, Hepatic venous pressure gradient (HVPG) was Diagnosis of sinusoidal obstruction syndrome (SOS) available in 30 out of the 50 suspected SOS cases and after hematopoietic cell transplantation (HCT) is based considered an additional diagnostic criterion. Acute re- on clinical criteria including weight gain, ascites, he- jection (AR) and antibody-­mediated rejection were spe- patomegaly, and jaundice.[1] However, clinical and cifically documented. histological features and prognosis of SOS after liver transplantation (LT) seem to differ from SOS after HCT.[2,3] We aimed to determine the characteristics and Histological evaluation outcomes of SOS after LT. Two liver pathologists (I.P. and J.P-­R.) blinded to clinical findings and previous pathological diagnosis indepen- PATI ENTS A ND M ETHO DS dently reviewed all biopsies (80 from 50 LTs) stained with hematoxylin-­eosin, Masson’s trichome, and reti- Possible SOS cases culin. Histologic assessment included parameters pro- posed by Rubbia-­Brandt et al.[4] Vascular injury and/or A case report form was sent to the 24 adult LT units perisinusoidal/central venous fibrosis were mandatory in Spain. Twenty units agreed to participate, including to establish SOS diagnosis. 12,941 LT procedures. Suspected SOS cases were identified as follows: In 43 cases, after SOS clinical suspicion, a local confirmatory Final diagnosis biopsy was available. In the remaining, SOS was estab- lished after identification of a biopsy coded with SOS di- After independent assessment, each case was jointly agnosis (n = 7). These biopsies were indicated based on reviewed to assign a final diagnosis. SOS diagnosis non-­SOS–­related clinical reasons or by protocol. was resolved in favor of histological diagnosis when In all cases, hepatic vein thrombosis and anasto- differences between clinical and histological evaluation motic stenosis were ruled out. appeared. Finally, 41 cases were diagnosed with SOS. Central confirmation of SOS Statistical analysis Fifty patients were identified as possible SOS cases Descriptive analysis was performed as appropriate ac- and centrally reevaluated as follows. cording to data distribution. Survival rates were calcu- lated and compared with the Kaplan-­Meier method and log-­rank test, respectively. Competing risk analysis was Clinical evaluation performed to estimate incidence of retransplantation considering death as a competing event. A multivari- All possible SOS cases were independently reviewed ate Fine and Gray regression analysis was performed, by 2 hepatologists (M.S. and A.C-­M.) blinded to histo- providing subhazard ratio (SHR) for SOS-­ specific logical findings. treatment. Abbreviations: AP, alkaline phosphatase; AR, acute rejection; CI, confidence interval; GGT, gamma-­glutamyl transferase; HCT, hematopoietic cell transplantation; HVPG, hepatic venous pressure gradient; IQR, interquartile range; LT, liver transplantation; PH, portal hypertension; SHR, subhazard ratio; SOS, sinusoidal obstruction syndrome; TIPS, transjugular intrahepatic portosystemic shunt. This is an open access article under the terms of the Creative Commons Attribution-­NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2022 The Authors. Liver Transplantation published by Wiley Periodicals LLC on behalf of American Association for the Study of Liver Diseases. Liver Transplantation. 2022;28:1257–1261.  wileyonlinelibrary.com/journal/lt | 1257 1258 |    BRIEF REPORT F I G U R E 1 (A) Sinusoidal dilatation around central vein (zone 3). (B) Perisinusoidal hemorrhage and sinusoidal dilatation. (C) Centrilobular vein fibrosis (hematoxylin-­eosin and Masson’s stain), obliterative fibrous lesion. (D) Centrilobular vein fibrosis, evolved obliterative lesion. (E) Peliosis hepatis. Irregular blood-­f illed cavities with no endothelial lining (hematoxylin-­eosin). (F) Nodular regenerative hyperplasia (hematoxylin-­eosin and reticulin stain). The nodularity results from zones of atrophic hepatocytes alternating with areas of normal-­sized hepatocytes R ESULTS present in only 39.0% of cases. At the time of histo- logical diagnosis, six patients had unspecific clinical Patientsʼ characteristics findings. HVPG was measured in 26 patients, being >6 mm The median age at diagnosis was 51 years (interquar- Hg in 25 (96.2%). A total of 21 (80.8%) patients had tile range [IQR] 44.0–­57.5), and 75.6% were male. clinically significant portal hypertension (PH; ie, HVPG The most frequent LT indications were advanced cir- >10 mm Hg). rhosis (51.2%) and hepatocellular carcinoma (31.7%). AR was diagnosed in 13 (31.7%) patients, 7 of them Tacrolimus was predominantly used during the first (53.9%) before SOS onset. No cases of antibody-­ month after LTs, in combination with either ster- mediated rejection were identified. oids (21.1%) or steroids and mycophenolate mofetil Most patients showed sinusoidal dilatation (97.6%; (36.8%). Figure 1A). Perisinusoidal hemorrhage (75.6%; Figure 1B) and centrilobular vein fibrosis (75.6%; Figure 1C,D) were also common. Peliosis (24.4%; SOS presentation Figure 1E) and nodular regenerative hyperplasia (26.8%; Figure 1F) were less frequent. No patient showed the classical triad of hepatomeg- The median time between LT and SOS onset was aly, ascites, and jaundice. Ascites was very frequent 2.43 months (IQR 1.64–­11.01). Many patients (70.7%) at the time of diagnosis (80.5%), whereas jaundice was were diagnosed within the first 6 months after LT, LIVER TRANSPLANTATION     | 1259 F I G U R E 2 Clinical presentation of SOS. (A) Interval between LT and SOS onset according to the presence of ascites. Bars represent mean time from liver transplantation to SOS in months. Error bars indicate the 95% confidence interval. (B) AP and GGT levels according to the interval between LT and SOS. Bars represent mean levels of AP and GGT. Error bars indicate the 95% confidence interval. (C) Comparison of HVPG according to the need of retransplantation. The box ranges from Q1 (the first quartile) to Q3 (the third quartile) of the distribution and the range represent the IQR. The median is indicated by a line across the box. The whiskers on box plots extend from Q1 and Q3 to the most extreme data points. (D) Cumulative incidence of retransplantation in the presence of competing risks in patients with SOS according to the type of treatment. Retransplantation as the event of interest and death as the competing event. (*p < 0.005) showing more frequently ascites (95.6% vs. 41.7%, p < 4–­16]). Unlike TIPS, defibrotide achieved histological 0.001; Figure 2A). Patients who presented SOS beyond resolution of SOS and PH in four patients (80.0%). 6 months after LT more frequently presented cholesta- Furthermore, defibrotide resolved clinical manifesta- sis (Figure 2B). tions in three patients (60.0%). No drug-­related ad- verse events were described. Six patients (14.6%) received anticoagulation (n = 3) or a mammalian target SOS management and outcomes of rapamycin inhibitor to minimize or stop tacrolimus (n = 3). The median follow-­up was 31.08 months (IQR 9.43–­ Six patients (14.6%) died (median follow-­up 31.08 71.63). Seventeen patients (41.5%) received only symp- months [range 0.37–­138.2]), three of them (50.0%) due tomatic therapy with diuretics, sodium restriction, and to SOS. Nine patients (22.0%) required retransplanta- paracentesis. Thirteen patients (31.7%) received a tion. SOS recurrence was not observed after retrans- transjugular intrahepatic portosystemic shunt (TIPS), plantation. Patients who required retransplantation with initial improvement in most of them (n = 9, 69.2%). had greater HVPG at diagnosis (Figure 2C). Moreover, Six out of seven cases with post-­TIPS HVPG measure- patients with an initial HVPG over 20 mm Hg required ments (85.7%) had a HVPG values in the normal range. retransplantation more frequently (88.9% vs. 11.1%; However, only one of the five cases who underwent a p = 0.02). The cumulative retransplantation-­free sur- control liver biopsy showed SOS resolution. Moreover, vival rates at 5 and 10 years were 65.3% and 58.1%, two patients (15.4%) who initially presented clinical res- respectively. olution required retransplantation after TIPS. Besides similar baseline characteristics, retrans- Five patients (12.2%) received defibrotide (median plantation was more frequently required in patients time between SOS onset and defibrotide: 14 days [IQR who received only symptomatic therapy, compared with 1260 |    BRIEF REPORT patients who received SOS-­specific treatment (62.5% Isabel Campos-­Varela5,2 vs. 22.2%; p = 0.06). Jordi Colmenero6,2 Patients who received SOS-­specific therapy showed Miguel Ángel Gómez-­Bravo7 a decreased risk of retransplantation (SHR for SOS Iago Justo8 specific treatment, 0.24; 95% confidence interval 0.05–­ Alejandra Otero9 1.15; p = 0.07) considering death as a competing event Esther Molina- ­Pérez10 (Figure 2D). Luisa González-­Diéguez11 Carme Baliellas12 Mario Romero- ­Cristobal1,2 D I SCUSS I O N Victoria Aguilera13 Lluís Castells5,2 We report the largest multicenter series of SOS cases Alba Díaz14,2 after LT aiming to ascertain its specific characteristics Luis Miguel Marín-­Gómez7 and clinical implications. Carmelo Loinaz8 We observed that the classic triad described for Rafael Bañares1,2 SOS after HCT[1] is extremely infrequent, with isolated Magdalena Salcedo1,2 ascites being the most common clinical presentation 1 after LT. Thus, consideration of SOS among the causes Hepatology and Liver Transplantation Unit, of graft dysfunction is essential to avoid misdiagnosis. Hospital General Universitario Gregorio Moreover, we have shown that the clinical presentation Marañón, Universidad Complutense, Madrid, pattern depends on the time of presentation. In fact, Spain 2 ascites is the most common presentation when SOS is Centro de Investigación Biomédica en Red diagnosed early after LT, while cholestasis is predomi- de Enfermedades Hepáticas y Digestivas nant in cases that appear later. (CIBERehd), Madrid, Spain 3 In our series, SOS was associated with poor out- Department of Pathology, Hospital General comes, including SOS-­related deaths and retransplan- Universitario Gregorio Marañón, Madrid, Spain 4 tations. Therefore, standardized diagnostic criteria for Department of Pathology, Hospital Universitario the specific setting of LT are needed. Considering our y Politécnico La Fe, Valencia, Spain 5 results, emphasizing the nonspecific clinical features, Liver Unit, Vall d’Hebron Hospital Universitari, histological evidence is essential to establish the diag- Vall d’Hebron Institut de Recerca (VHIR), nosis after LT. According to the proposed pathogenetic Vall d’Hebron Barcelona Hospital Campus, SOS mechanisms, vascular injury and/or perisinusoi- Universitat Autònoma de Barcelona, Barcelona, dal/central venous fibrosis should be considered the Spain landmark to establish SOS diagnosis.[5] 6 Liver Transplant Unit, Hospital Clínic, IDIBAPS, Another important finding is the clear influence of University of Barcelona, Barcelona, Spain 7 PH in SOS survival and need for retransplantation. Liver Transplant Unit, Hospital Virgen del Rocío, We found a threshold HVPG value (>20 mm Hg) that Seville, Spain 8 identifies a population at greater risk. Thus, inclusion of Department of Hepatology/HPB-­Surgery/ HVPG measurement should be part of the diagnostic Transplantation, Hospital Universitario 12 de workup. Octubre, Madrid, Spain 9 Finally, we have found that patients treated only Liver Transplant Unit, Hospital de A Coruña, A with symptomatic therapy are more likely to require re- Coruña, Spain 10 transplantation after SOS diagnosis. Therefore, SOS-­ Department of Liver Transplantation, Hospital specific therapies, such as defibrotide or TIPS, should Universitario de Santiago, Santiago de be considered. Compostela, Spain 11 In conclusion, this case series provides valuable in- Liver Unit and Division of Gastroenterology and formation helping in the diagnosis and management of Hepatology, Hospital Universitario Central de SOS after LT. Asturias, Oviedo, Spain 12 Liver Transplant Unit, Hospital Universitari de CONFLICT OF INTEREST Bellvitge, IDIBELL, L’Hospitalet de Llobregat, Jordi Colmenero consults for and is on the speakers' Barcelona, Spain 13 bureau for Chiesi. He is on the speakers' bureau for Department of Hepatology and Liver Astellas. The other authors do not have any conflicts. Transplantation, Hospital Universitario y Politécnico La Fe, Valencia, Spain Aránzazu Caballero- ­Marcos1,2 14 Department of Pathology, Hospital Clínic, Isabel Peligros3 IDIBAPS, University of Barcelona, Judith Pérez-­Rojas4 Barcelona, Spain LIVER TRANSPLANTATION     | 1261 Correspondence Victoria Aguilera https://orcid. Magdalena Salcedo, Hepatology and Liver org/0000-0001-6492-0357 Transplantation Unit, Hospital General Lluís Castells https://orcid. Universitario Gregorio Marañón, Universidad org/0000-0002-6672-5931 Complutense, C/ Dr. Esquerdo, 46. CP 28007, Alba Díaz https://orcid.org/0000-0002-0456-7085 Madrid, Spain. Luis Miguel Marín-­Gómez https://orcid. Email: magdalena.salcedo@icloud.com org/0000-0002-6241-0326 Carmelo Loinaz https://orcid. ORCID org/0000-0002-1873-0568 Aránzazu Caballero- ­Marcos https://orcid. Rafael Bañares https://orcid. org/0000-0001-8116-213X org/0000-0002-0412-8437 Isabel Peligros https://orcid. Magdalena Salcedo https://orcid. org/0000-0002-1271-2780 org/0000-0002-1239-5746 Judith Pérez-­Rojas https://orcid. org/0000-0001-9558-654X REFERENCES Isabel Campos-­Varela https://orcid. 1. Mohty M, Malard F, Abecassis M, Aerts E, Alaskar AS, Aljurf M, org/0000-0001-6597-3151 et al. Sinusoidal obstruction syndrome/veno-­occlusive disease: current situation and perspectives—­a position statement from Jordi Colmenero https://orcid. the European Society for Blood and Marrow Transplantation org/0000-0001-6024-8479 (EBMT). Bone Marrow Transplant. 2015;50:781–­9. Miguel Ángel Gómez-­Bravo https://orcid. 2. Sebagh M, Azoulay D, Roche B, Hoti E, Karam V, Teicher E, org/0000-0002-5155-3508 et al. Significance of isolated hepatic veno-­occlusive disease/ Iago Justo https://orcid.org/0000-0002-0553-5835 sinusoidal obstruction syndrome after liver transplantation. Alejandra Otero https://orcid. Liver Transpl. 2011;17:798–­808. 3. Sebagh M, Debette M, Samuel D, Emile J-­F, Falissard B, Cailliez org/0000-0002-9016-3705 V, et al. “Silent” presentation of veno-­occlusive disease after Esther Molina-­Pérez https://orcid. liver transplantation as part of the process of cellular rejection org/0000-0002-6951-3600 with endothelial predilection. Hepatology. 1999;30:1144–­50. Luisa González- ­Diéguez https://orcid. 4. Rubbia-­Brandt L, Lauwers GY, Wang H, Majno PE, Tanabe K, org/0000-0003-1098-3891 Zhu AX, et al. Sinusoidal obstruction syndrome and nodular re- generative hyperplasia are frequent oxaliplatin-­associated liver Carme Baliellas https://orcid. lesions and partially prevented by bevacizumab in patients with org/0000-0002-3641-6484 hepatic colorectal metastasis. Histopathology. 2010;56:430–­9. Mario Romero- ­Cristobal https://orcid. 5. Sinusoidal R-­ B, Syndrome O. Sinusoidal obstruction syn- org/0000-0003-1633-2862 drome. Clin Liver Dis. 2010;14:651–­68.