Risk of venous thromboembolism in splenectomized patients compared with the general population and appendectomized patients: a 10-year nationwide cohort study.

Abstract

Splenectomy has been associated with hemorrhagic and other peri‐ or postoperative complications [1Winslow E.R. Brunt L.M. Perioperative outcomes of laparoscopic versus open splenectomy: a meta‐analysis with an emphasis on complications.Surgery. 2003; 134: 647-53Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar], and splenectomized patients are at lifelong increased risk of severe bacterial infections and mortality [2Kyaw M.H. Holmes E.M. Toolis F. Wayne B. Chalmers J. Jones I.G. Campbell H. Evaluation of severe infection and survival after splenectomy.Am J Med. 2006; 119: 276-7Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, 3Cadili A. De Gara C. Complications of splenectomy.Am J Med. 2008; 121: 371-5Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar, 4Thomsen R.W. Schoonen W.M. Farkas D.K. Riis A. Jacobsen J. Fryzek J.P. Sørensen H.T. Risk for hospital contact with infection in patients with splenectomy: a population‐based cohort study.Ann Intern Med. 2009; 151: 546-55Crossref PubMed Google Scholar]. Other potential complications include venous thromboembolism (VTE) [3Cadili A. De Gara C. Complications of splenectomy.Am J Med. 2008; 121: 371-5Abstract Full Text Full Text PDF PubMed Scopus (114) Google Scholar], including portal vein thrombosis usually originating in the splenic vein [5Chaffanjon P.C. Brichon P.Y. Ranchoup Y. Gressin R. Sotto J.J. Portal vein thrombosis following splenectomy for hematologic disease: prospective study with Doppler color flow imaging.World J Surg. 1998; 22: 1082-6Crossref PubMed Scopus (113) Google Scholar, 6Fujita F. Lyass S. Otsuka K. Giordano L. Rosenbaum D.L. Khalili T.M. Phillips E.H. Portal vein thrombosis following splenectomy: identification of risk factors.Am Surg. 2003; 69: 951-6Crossref PubMed Google Scholar, 7Hassn A.M. Al Fallouji M.A. Ouf T.I. Saad R. Portal vein thrombosis following splenectomy.Br J Surg. 2000; 87: 362-73Crossref PubMed Google Scholar, 8Ikeda M. Sekimoto M. Takiguchi S. Kubota M. Ikenaga M. Yamamoto H. Fujiwara Y. Ohue M. Yasuda T. Imamura H. Tatsuta M. Yano M. Furukawa H. Monden M. High incidence of thrombosis of the portal venous system after laparoscopic splenectomy: a prospective study with contrast‐enhanced CT scan.Ann Surg. 2005; 241: 208-16Crossref PubMed Scopus (0) Google Scholar, 9Pietrabissa A. Moretto C. Antonelli G. Morelli L. Marciano E. Mosca F. Thrombosis in the portal venous system after elective laparoscopic splenectomy.Surg Endosc. 2004; 18: 1140-3Crossref PubMed Google Scholar, 10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar, 11Van’t Riet M. Burger J.W. Van Muiswinkel J.M. Kazemier G. Schipperus M.R. Bonjer H.J. Diagnosis and treatment of portal vein thrombosis following splenectomy.Br J Surg. 2000; 87: 1229-33Crossref PubMed Scopus (0) Google Scholar, 12Winslow E.R. Brunt L.M. Drebin J.A. Soper N.J. Klingensmith M.E. Portal vein thrombosis after splenectomy.Am J Surg. 2002; 184: 631-5Abstract Full Text Full Text PDF PubMed Scopus (158) Google Scholar]. Postoperative deep vein thrombosis (DVT) in the lower leg and pulmonary embolism (PE) have also been reported [6Fujita F. Lyass S. Otsuka K. Giordano L. Rosenbaum D.L. Khalili T.M. Phillips E.H. Portal vein thrombosis following splenectomy: identification of risk factors.Am Surg. 2003; 69: 951-6Crossref PubMed Google Scholar, 10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar]. However, epidemiological data on the magnitude and duration of increased VTE risk in splenectomized patients are sparse [10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar, 13Parker III, H.H. Bynoe R.P. Nottingham J.M. Thrombosis of the portal venous system after splenectomy for trauma.J Trauma. 2003; 54: 193-6Crossref PubMed Scopus (0) Google Scholar, 14Peacock A.J. Pulmonary hypertension after splenectomy: a consequence of loss of the splenic filter or is there something more?.Thorax. 2005; 60: 983-4Crossref PubMed Scopus (0) Google Scholar], and few splenectomy studies have followed patients beyond 1 year [2Kyaw M.H. Holmes E.M. Toolis F. Wayne B. Chalmers J. Jones I.G. Campbell H. Evaluation of severe infection and survival after splenectomy.Am J Med. 2006; 119: 276-7Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar, 4Thomsen R.W. Schoonen W.M. Farkas D.K. Riis A. Jacobsen J. Fryzek J.P. Sørensen H.T. Risk for hospital contact with infection in patients with splenectomy: a population‐based cohort study.Ann Intern Med. 2009; 151: 546-55Crossref PubMed Google Scholar, 15Bisharat N. Omari H. Lavi I. Raz R. Risk of infection and death among post‐splenectomy patients.J Infect. 2001; 43: 182-6Abstract Full Text PDF PubMed Scopus (0) Google Scholar, 16Cullingford G.L. Watkins D.N. Watts A.D. Mallon D.F. Severe late postsplenectomy infection.Br J Surg. 1991; 78: 716-21Crossref PubMed Scopus (250) Google Scholar, 17Ejstrud P. Kristensen B. Hansen J.B. Madsen K.M. Schonheyder H.C. Sorensen H.T. Risk and patterns of bacteraemia after splenectomy: a population‐based study.Scand J Infect Dis. 2000; 32: 521-5Crossref PubMed Google Scholar, 18Schwartz P.E. Sterioff S. Mucha P. Melton III, L.J. Offord K.P. Postsplenectomy sepsis and mortality in adults.JAMA. 1982; 248: 2279-83Crossref PubMed Scopus (251) Google Scholar]. We conducted a nationwide, population‐based assessment of risk of VTE‐related hospital contacts among splenectomized patients compared with Denmark’s general population and appendectomized patients. As described in detail elsewhere [4Thomsen R.W. Schoonen W.M. Farkas D.K. Riis A. Jacobsen J. Fryzek J.P. Sørensen H.T. Risk for hospital contact with infection in patients with splenectomy: a population‐based cohort study.Ann Intern Med. 2009; 151: 546-55Crossref PubMed Google Scholar], we used data from the Danish National Patient Registry (NPR) to identify all Danish residents with a first splenectomy performed between 1 January 1996 and 31 December 2005. We classified splenectomy patients into eight groups according to indication, using the following hierarchy [19Mellemkjoer L. Olsen J.H. Linet M.S. Gridley G. McLaughlin J.K. Cancer risk after splenectomy.Cancer. 1995; 75: 577-83Crossref PubMed Scopus (105) Google Scholar]: (i) traumatic splenic rupture; (ii) immune thrombocytopenic purpura (ITP); (iii) other/unspecific thrombocytopenia; (iv) hematopoietic cancer; (v) hereditary hemolytic anemia; (vi) abdominal cancer; (vii) splenomegaly/other splenic diseases only; and (viii) none of these diagnoses. From the Danish Civil Registration System (CRS), a nationwide registry of all Danish residents [20Frank L. Epidemiology. When an entire country is a cohort.Science. 2000; 287: 2398-9Crossref PubMed Scopus (721) Google Scholar], we randomly chose 10 general population comparison subjects for each splenectomy patient, matched by age and sex. In addition, using the CRS and NPR we chose for each splenectomy patient up to five matched comparison patients who had an appendectomy. Each comparison subject had to be alive with no splenectomy history before or on the index date (i.e. the date the corresponding matched patient underwent splenectomy). All in‐ and outpatient hospital contacts involving a VTE diagnosis after the index date among cohort members were retrieved from the CRS and the NPR. We focused on any VTE event overall, and on the following subcategories: deep vein thrombosis or phlebitis (DVT), pulmonary embolism and infarction (PE), portal and/or splenic vein thrombosis (PSVT), and other VTE, including Budd‐Chiari syndrome and embolism/thrombosis of other/unspecified veins. To control for potential confounding factors we retrieved data on other medical conditions diagnosed before the index date, including malignancy, cardiovascular disease (previous myocardial infarction, congestive heart failure, peripheral vascular disease, cerebrovascular disease, or hypertension), gastrointestinal disease (inflammatory bowel disease, peptic ulcer disease, pancreatitis, or liver disease), chronic pulmonary disease, alcoholism, diabetes, connective tissue disease, obesity, renal disease, AIDS, hemiplegy, and dementia. We assessed the association between splenectomy and subsequent VTE events, overall and separately by splenectomy indication, compared with the two comparison cohorts within 90 days, 91–365 days, and more than 365 days after the index date. Because the exact date of a VTE event during the splenectomy hospitalization is unavailable in the NRP [17Ejstrud P. Kristensen B. Hansen J.B. Madsen K.M. Schonheyder H.C. Sorensen H.T. Risk and patterns of bacteraemia after splenectomy: a population‐based study.Scand J Infect Dis. 2000; 32: 521-5Crossref PubMed Google Scholar, 21Wiseman J. Brown C.V. Weng J. Salim A. Rhee P. Demetriades D. Splenectomy for trauma increases the rate of early postoperative infections.Am Surg. 2006; 72: 947-50Crossref PubMed Google Scholar], we estimated 90‐day odds ratios (ORs) with 95% confidence intervals (CIs), using logistic regression, as the measure of 90‐day relative risk (RR). All VTE diagnoses during initial hospitalizations that ended within 90 days, or new hospital admissions beginning within 90 days, were included. Cox’s proportional hazards regression was used to compute hazard ratios with 95% CIs as measures of RR of a new hospital contact with VTE during the follow‐up periods of 91–365 days and 365+ days. All models were adjusted for age, gender and the medical conditions defined above, diagnosed before the index date. We used sas software (Version 9.1, SAS Institute, Cary, NC, USA). The Danish Data Protection Agency and Aarhus University Hospital Registry Board approved the study. We identified 3812 splenectomized individuals, 38 120 general population members, and 16 962 appendectomy cohort members. Splenectomy patients had a higher burden of comorbidity than the general population or appendectomy comparisons, particularly malignancies, cardiovascular disease and gastrointestinal disorders. Within the first 90 days postsplenectomy, any VTE event affected 71 (1.9%) of splenectomy patients vs. 25 (0.1%) of general population comparisons (adjusted OR = 32.6, 95% CI, 18.3–58.0) and vs. 95 (0.6%) of appendectomized comparisons (adjusted OR = 3.2, 95% CI, 2.3–4.6) (Table 1). We found a 19.8‐fold (95% CI, 8.8–44.7) and 2.3‐fold (95% CI, 1.3–4.1) higher risk of DVT and a 32.6‐fold (95% CI, 13.9–76.3) and 3.2‐fold (95% CI, 1.8–5.5) higher risk of PE in splenectomized patients compared with the general population and appendectomy patients, respectively. ORs were even higher for PSVT but could not be meaningfully estimated due to this event being rare among comparisons (Table 1). During the 91–365 days following splenectomy, the adjusted RR of any VTE among the splenectomized patients was 7.1 (95% CI, 4.4–11.3) compared with the general population and 2.8 (95% CI, 1.7–4.4) compared with the appendectomized patients. More than 1 year postsplenectomy, the adjusted RR for any VTE was 3.4 compared with the general population and 3.2 compared with the appendectomy cohort. Intermediate and long‐term PVST risk was greatly increased but risk estimates were imprecise (Table 1).Table 1Risks of hospital contact involving venous thromboembolism (VTE) overall and stratified by VTE type in splenectomized patients and the general population and appendectomy comparison subjects, according to time since splenectomySplenectomized patients (n = 3812)Population comparison subjects (n = 38 120)Adjusted relative risk* (95% CI)Appendectomy comparison subjects (n = 16 962)Adjusted relative risk* (95% CI)90 days follow‐up after splenectomyAny VTE involving hospital contact, n (%)71 (1.86)25 (0.07)32.6 (18.3–58.0)95 (0.56)3.2 (2.3–4.6)DVT, n (%)24 (0.63)16 (0.04)19.8 (8.7–44.7)50 (0.29)2.3 (1.3–4.1)PE, n (%)28 (0.73)8 (0.02)32.6 (13.9–76.3)45 (0.27)3.2 (1.8–5.5)PSVT, n (%)15 (0.39)0 (0.00)N/A2 (0.01)N/A [crude 37.5 (8.6–164.0)]Other VTE, n (%)11 (0.29)2 (0.01)43.1 (4.7–396.2)2 (0.01)N/A [crude 26.1 (5.8–118.2)]91–365 days follow‐up after splenectomyAny VTE involving hospital contact, n (per 100 PY)42 (1.90)74 (0.26)7.1 (4.4–11.3)65 (0.53)2.8 (1.7–4.4)DVT, n (per 100 PY)22 (0.99)45 (0.16)5.7 (3.0–10.7)44 (0.36)2.2 (1.2–4.1)PE, n (per 100 PY)9 (0.40)26 (0.09)5.1 (2.0–13.5)21 (0.17)1.3 (0.5–3.5)PSVT, n (per 100 PY)10 (0.45)1 (0.00)N/A1 (0.01)N/AOther VTE, n (per 100 PY)3 (0.13)5 (0.02)2.1 (0.1–30.1)4 (0.03)N/A [crude 6.5 (1.1–39.3)More than 365 days follow‐up after splenectomyAny VTE involving hospital contact, n (per 100 PY)118 (1.03)545 (0.31)3.4 (2.7–4.3)264 (0.35)3.2 (2.4–4.2)DVT, n (per 100 PY)78 (0.68)347 (0.20)3.7 (2.8–4.9)170 (0.23)3.4 (2.4–4.8)PE, n (per 100 PY)38 (0.33)199 (0.11)3.5 (2.3–5.3)83 (0.11)3.3 (2.0–5.4)PSVT, n (per 100 PY)12 (0.10)3 (0.00)N/A [crude 116.3 (15.1–894.5)]4 (0.01)N/A [crude 20.0 (5.6–70.9)]Other VTE, n (per 100 PY)11 (0.09)38 (0.02)3.1 (1.2–7.9)24 (0.03)2.6 (0.9–7.1)VTE, venous thromboembolism; DVT, phlebitis or deep venous thromboembolism of the lower limb; PE, pulmonary embolism; PSVT, portal or splenic vein thrombosis; PY, person‐years. N/A, not applicable. N/A listed when the number of events was too small to obtain a meaningful risk measure or adjustment.*Relative risk estimated as odds ratio of VTE within 90 days, and as incidence rate ratio of VTE in the time periods of 91–365 days and more than 365 days after splenectomy. Relative risk estimates adjusted by age, sex, and comorbid conditions present at the time of splenectomy. Open table in a new tab VTE, venous thromboembolism; DVT, phlebitis or deep venous thromboembolism of the lower limb; PE, pulmonary embolism; PSVT, portal or splenic vein thrombosis; PY, person‐years. N/A, not applicable. N/A listed when the number of events was too small to obtain a meaningful risk measure or adjustment. *Relative risk estimated as odds ratio of VTE within 90 days, and as incidence rate ratio of VTE in the time periods of 91–365 days and more than 365 days after splenectomy. Relative risk estimates adjusted by age, sex, and comorbid conditions present at the time of splenectomy. The most common medical indications for splenectomy were traumatic rupture of the spleen (20.1%) and abdominal cancer (19.3%). Within 90 days after splenectomy, the adjusted ORs of VTE were 16‐fold to 78‐fold higher in all eight indication subgroups than in general population comparisons. Ninety‐day ORs compared with appendectomized patients ranged from 2.0 in abdominal cancer patients to 7.6 in hematopoietic cancer patients with splenectomy. Long‐term VTE RRs were highest in patients splenectomized due to malignant hematological disorders, hemolytic anemia or splenomegaly as the only recorded indication (365+ days RRs ranged from 8 to 22 vs. the general population and from 6 to 11 vs. appendectomized patients in these three groups). Among the 765 relatively young and previously healthy individuals who had suffered splenic trauma (82% were younger than 60 years), the 365+ days risk of VTE remained 3.1‐fold higher (95% CI, 1.8–5.4) than in their age‐matched population comparisons and 3.7‐fold higher (95% CI, 1.8–7.5) than in their age‐matched appendectomized comparisons. For patients with ITP, the corresponding 365+ days risks were 2.7‐fold (95% CI, 1.1–6.3) and 2.6‐fold (95% CI, 0.9–7.1) increased. Our estimates agree with previous observations of a high incidence of VTE events after splenectomy. In a Dutch University Hospital, nine (1.6%) of 563 splenectomized patients developed symptomatic portal vein thrombosis [11Van’t Riet M. Burger J.W. Van Muiswinkel J.M. Kazemier G. Schipperus M.R. Bonjer H.J. Diagnosis and treatment of portal vein thrombosis following splenectomy.Br J Surg. 2000; 87: 1229-33Crossref PubMed Scopus (0) Google Scholar] compared with six (0.9%) of 688 splenectomized patients in a California setting [6Fujita F. Lyass S. Otsuka K. Giordano L. Rosenbaum D.L. Khalili T.M. Phillips E.H. Portal vein thrombosis following splenectomy: identification of risk factors.Am Surg. 2003; 69: 951-6Crossref PubMed Google Scholar]. In some smaller case series, 7–15% of patients suffered symptomatic portal vein thrombosis postsplenectomy [7Hassn A.M. Al Fallouji M.A. Ouf T.I. Saad R. Portal vein thrombosis following splenectomy.Br J Surg. 2000; 87: 362-73Crossref PubMed Google Scholar, 8Ikeda M. Sekimoto M. Takiguchi S. Kubota M. Ikenaga M. Yamamoto H. Fujiwara Y. Ohue M. Yasuda T. Imamura H. Tatsuta M. Yano M. Furukawa H. Monden M. High incidence of thrombosis of the portal venous system after laparoscopic splenectomy: a prospective study with contrast‐enhanced CT scan.Ann Surg. 2005; 241: 208-16Crossref PubMed Scopus (0) Google Scholar, 9Pietrabissa A. Moretto C. Antonelli G. Morelli L. Marciano E. Mosca F. Thrombosis in the portal venous system after elective laparoscopic splenectomy.Surg Endosc. 2004; 18: 1140-3Crossref PubMed Google Scholar, 10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar, 12Winslow E.R. Brunt L.M. Drebin J.A. Soper N.J. Klingensmith M.E. Portal vein thrombosis after splenectomy.Am J Surg. 2002; 184: 631-5Abstract Full Text Full Text PDF PubMed Scopus (158) Google Scholar]. Three studies prospectively screened splenectomized patients for thrombosis in the splenic and/or portal vein system with Doppler ultrasound [5Chaffanjon P.C. Brichon P.Y. Ranchoup Y. Gressin R. Sotto J.J. Portal vein thrombosis following splenectomy for hematologic disease: prospective study with Doppler color flow imaging.World J Surg. 1998; 22: 1082-6Crossref PubMed Scopus (113) Google Scholar, 9Pietrabissa A. Moretto C. Antonelli G. Morelli L. Marciano E. Mosca F. Thrombosis in the portal venous system after elective laparoscopic splenectomy.Surg Endosc. 2004; 18: 1140-3Crossref PubMed Google Scholar] or contrast‐enhanced computed tomography (CT) scans [8Ikeda M. Sekimoto M. Takiguchi S. Kubota M. Ikenaga M. Yamamoto H. Fujiwara Y. Ohue M. Yasuda T. Imamura H. Tatsuta M. Yano M. Furukawa H. Monden M. High incidence of thrombosis of the portal venous system after laparoscopic splenectomy: a prospective study with contrast‐enhanced CT scan.Ann Surg. 2005; 241: 208-16Crossref PubMed Scopus (0) Google Scholar]. They found that 8–37% of all patients developed PSVT, of which 2–15% had associated symptoms [5Chaffanjon P.C. Brichon P.Y. Ranchoup Y. Gressin R. Sotto J.J. Portal vein thrombosis following splenectomy for hematologic disease: prospective study with Doppler color flow imaging.World J Surg. 1998; 22: 1082-6Crossref PubMed Scopus (113) Google Scholar, 8Ikeda M. Sekimoto M. Takiguchi S. Kubota M. Ikenaga M. Yamamoto H. Fujiwara Y. Ohue M. Yasuda T. Imamura H. Tatsuta M. Yano M. Furukawa H. Monden M. High incidence of thrombosis of the portal venous system after laparoscopic splenectomy: a prospective study with contrast‐enhanced CT scan.Ann Surg. 2005; 241: 208-16Crossref PubMed Scopus (0) Google Scholar, 9Pietrabissa A. Moretto C. Antonelli G. Morelli L. Marciano E. Mosca F. Thrombosis in the portal venous system after elective laparoscopic splenectomy.Surg Endosc. 2004; 18: 1140-3Crossref PubMed Google Scholar]. It is unknown how much lack of PSVT coding in database studies like ours or publication bias in smaller case series contribute to these discrepancies. Previous data on DVT or PE risk are sparse. A Swedish case series found that 2 (3%) of 69 splenectomized patients developed DVT, both with ITP [10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar]. The mechanisms underlying the association between splenectomy and VTE are not fully understood. Surgery itself raises VTE risk, as also suggested by the less markedly increased early risks we found compared with appendectomized patients. Most splenectomy‐associated PSVTs seem to occur in patients with massive splenomegaly, malignant hematological disorders (particularly myelopoliferative disorders), high platelet counts or hemolysis. Splenectomy may lead to both thrombocytosis and an increased number of damaged circulating red cells, in turn leading to hypercoagulability [22Boxer M.A. Braun J. Ellman L. Thromboembolic risk of postsplenectomy thrombocytosis.Arch Surg. 1978; 113: 808-9Crossref PubMed Google Scholar, 23Cappellini M.D. Robbiolo L. Bottasso B.M. Coppola R. Fiorelli G. Mannucci A.P. Venous thromboembolism and hypercoagulability in splenectomized patients with thalassaemia intermedia.Br J Haematol. 2000; 111: 467-73Crossref PubMed Scopus (212) Google Scholar]. However, indications for splenectomy such as myelopoliferative disorders or hemolytic anemia may themselves cause hypercoaguability through increased hematocrit or hemolysis of red cells [14Peacock A.J. Pulmonary hypertension after splenectomy: a consequence of loss of the splenic filter or is there something more?.Thorax. 2005; 60: 983-4Crossref PubMed Scopus (0) Google Scholar, 24Sorensen H.T. Mellemkjaer L. Steffensen F.H. Olsen J.H. Nielsen G.L. The risk of a diagnosis of cancer after primary deep venous thrombosis or pulmonary embolism.N Engl J Med. 1998; 338: 1169-73Crossref PubMed Scopus (562) Google Scholar]. Nevertheless, we also found clearly increased short‐ and long‐term rates of VTE among patients with traumatic splenic rupture. Trauma, orthopedic surgery and immobility due to post‐accident sequelae may have affected VTE risk in these patients. Also, because of underlying indications for splenectomy, patients may experience VTE due to more frequent hospitalizations with associated immobilization. Within an enlarged spleen the splenic vein may function as a ‘cul‐de‐sac’ where thromboses can grow and spread [5Chaffanjon P.C. Brichon P.Y. Ranchoup Y. Gressin R. Sotto J.J. Portal vein thrombosis following splenectomy for hematologic disease: prospective study with Doppler color flow imaging.World J Surg. 1998; 22: 1082-6Crossref PubMed Scopus (113) Google Scholar, 10Svensson M. Wiren M. Kimby E. Hagglund H. Portal vein thrombosis is a common complication following splenectomy in patients with malignant haematological diseases.Eur J Haematol. 2006; 77: 203-9Crossref PubMed Scopus (33) Google Scholar], consistent with our observation of the highest VTE and PVST risk in patients with splenomegaly. The surgical procedure data used for splenectomy as well as hospital discharge diagnoses of PE and DVT have documented high validity [25Kniffin Jr, WD. Baron J.A. Barrett J. Birkmeyer J.D. Anderson Jr, F.A. The epidemiology of diagnosed pulmonary embolism and deep venous thrombosis in the elderly.Arch Intern Med. 1994; 154: 861-6Crossref PubMed Google Scholar, 26Larsen T.B. Johnsen S.P. Moller C.I. Larsen H. Sorensen H.T. A review of medical records and discharge summary data found moderate to high predictive values of discharge diagnoses of venous thromboembolism during pregnancy and postpartum.J Clin Epidemiol. 2005; 58: 316-9Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]. Increased surveillance of splenectomized patients may have led to overestimation of VTE risk, yet events such as PE or portal vein thrombosis are typically accompanied by severe symptoms requiring hospitalization. Although we were able to control for a wide range of known confounders [27Goldhaber S.Z. Pulmonary embolism.Lancet. 2004; 363: 1295-305Abstract Full Text Full Text PDF PubMed Scopus (372) Google Scholar, 28Kyrle P.A. Eichinger S. Deep vein thrombosis.Lancet. 2005; 365: 1163-74Abstract Full Text Full Text PDF PubMed Scopus (420) Google Scholar, 29Sorensen H.T. Horvath‐Puho E. Pedersen L. Baron J.A. Prandoni P. Venous thromboembolism and subsequent hospitalisation due to acute arterial cardiovascular events: a 20‐year cohort study.Lancet. 2007; 370: 1773-9Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar], we lacked data on some factors, including smoking, alcohol intake and body mass index, or exact information on immobility, and use of contraceptives or HRT [27Goldhaber S.Z. Pulmonary embolism.Lancet. 2004; 363: 1295-305Abstract Full Text Full Text PDF PubMed Scopus (372) Google Scholar, 28Kyrle P.A. Eichinger S. Deep vein thrombosis.Lancet. 2005; 365: 1163-74Abstract Full Text Full Text PDF PubMed Scopus (420) Google Scholar]. We also lacked data on perioperative thrombo‐prophylaxis. By 2005, 96% of surgical departments in Denmark routinely used peri‐ and postoperative thrombo‐prophylaxis with low molecular weight heparin (LMWH) for major surgery [30Schmidt L.M. Foli‐Andersen N.J. Rasmussen H.M. Wille‐Jorgensen P.A. Thrombo‐prophylaxis in Danish surgical departments. Status 2005 and 25 years’ development.Ugeskr Laeger. 2008; 170: 947-51PubMed Google Scholar]. This nationwide population‐based study indicates that splenectomized patients are at increased short‐ and long‐term risk of VTE involving hospital contact. Relative risks of VTE remain 3‐fold elevated a year after the procedure, compared with the general population or appendectomized patients. As VTE events may be associated with long‐term sequelae such as pulmonary hypertension [14Peacock A.J. Pulmonary hypertension after splenectomy: a consequence of loss of the splenic filter or is there something more?.Thorax. 2005; 60: 983-4Crossref PubMed Scopus (0) Google Scholar] and mortality [27Goldhaber S.Z. Pulmonary embolism.Lancet. 2004; 363: 1295-305Abstract Full Text Full Text PDF PubMed Scopus (372) Google Scholar], patients who have undergone splenectomy should be carefully monitored. Clinical studies are needed to better understand which patients are at risk of VTE and who may benefit from prophylactic anticoagulant treatment. W.M. Schoonen and J.P. Fryzek are employees of and own stock in Amgen. The other authors state that they have no conflict of interest. The authors state that they have no conflict of interest. This study was partly supported by: a research grant from Amgen Inc., USA, to Aarhus University; the Clinical Epidemiological Research Foundation at Aarhus University; and the Karen Elise Jensens Foundation, Denmark.