Abstract
A 48-year-old woman presented with severe abdominal pain, bilious vomiting and bloody diarrhoea for 1 day. On examination, she was haemodynamically unstable, febrile and clinically had an acute surgical abdomen. She had markedly raised inflammatory markers, neutrophils and deranged renal function. A CT abdominal scan revealed severe colitis and thickening throughout the length of the colon. The patient was stabilised and underwent emergency laparotomy resulting in total colectomy and end ileostomy formation. Postoperatively, she required several units of human albumin solution, red blood cell transfusions and octaplex (prothrombin complex) to prevent further bleeding. An inpatient haematology review revealed a hypocomplementaemia (C3/C4), low immunoglobulin (IgG, IgM, IgA) and peripheral blood films revealed schistocytosis indicating microangiopathic haemolytic anaemia. Bowel histology supported this, demonstrating circumferential lymphocytic phlebitis with thrombi and mucosal haemorrhage, necrosis and ulceration. The patient went on to suffer multiple ischaemic strokes before undergoing plasmapheresis, subsequent rehabilitation and making a successful recovery.
Keywords: haematology (incl blood transfusion), GI bleeding, haematology (drugs and medicines), surgery, general surgery
Background
Thrombotic microangiopathies (TMAs) are a collection of rare, potentially life-threatening conditions that cause thrombotic occlusion of small vessels. Untreated mortality in some forms can be up to 90%.1 The clinical syndrome of organ dysfunction, microangiopathy haemolytic anaemia and thrombocytopenia rapidly ensues. Prompt and accurate diagnosis is complicated by the differing clinical signs and symptoms associated with each condition, delaying critical interventions and supportive therapies—in this case leading to neurological complications. We therefore aim to raise awareness of these potentially fatal conditions so that appropriate investigation and management can be initiated without delay. TMA should be suspected in patients found to have thrombocytopaenia and microangiopathic haemolytic anaemia without other identifiable causes. The British Society for Haematology suggest that diagnosis should be based on clinical history, examination and blood film but acknowledge that diagnosis of thrombotic thrombocytopaenic purpura (TTP) and other subtypes of TMA remains challenging.1 This case demonstrates the complexity of TMA and highlights that management should focus on a multidisciplinary approach involving haematologists, physicians and surgeons and should be treated as a medical emergency.
Case presentation
A 48-year-old female teacher presented acutely unwell after work to the surgical assessment unit with a 1 day history of severe central abdominal pain, bloody diarrhoea, copious bilious vomiting and fever. The patient reported feeling confused and disorientated, lethargic and complained of a headache. She denied any weight loss and changes in appetite, any urinary symptoms and any possibility of pregnancy or ingestion of contaminated food or water. There was a prior medical history of pancreatitis (unknown aetiology), urticaria, external haemorrhoids and an uneventful caesarean section delivery. Significant family history included bowel cancer. She was not on any regular medications, nor any alternative therapies or supplements, but had an intrauterine contraceptive device. The patient maintained a balanced diet, did not smoke nor use recreational drugs. She often travelled and enjoyed hiking as a pastime. On initial inspection, the patient had decreased consciousness and confusion (Glasgow Coma Scale rating of 13/15) and was tachypnoeic, hypotensive and tachycardic on initial observations. She was also nauseous and vomiting. The abdomen was tense, exquisitely tender throughout and distended on palpation. On digital rectal examination, there was significant per rectum bleeding and clot formation. Neurological examination was limited due to the patient’s decreased consciousness; however, gross examination of cranial nerves was unremarkable and there was no apparent central or peripheral motor or sensory dysfunction. Cardiovascular examination revealed a sinus tachycardia, hypotension and faint peripheral pulses.
Investigations
At presentation, a beta human chorionic gonadotrophin test was negative and urine analysis was importantly positive for protein (3+) and haemoglobin (3+), however, negative for nitrites and leucocytes. A urine drug screen was negative.
Initial blood tests showed a leucocytosis (white cell count 27 500 cells/mm3, neutrophils 25 300 cells/mm3), severe thrombocytopaenia (platelets 22 000 cells/mm3) and a moderately deranged renal function (estimated glomerular filtration rate 49 mL/min/1.73 m2, previously >90). Initially, the patient had a haemoglobin of 205 g/L (likely raised due to dehydration); however, this deteriorated critically to 64 g/L by day 4, indicating a severe haemolytic process.
An urgent contrast CT abdomen and pelvis scan revealed extensive ischaemic colitis and thickening within the entirety of the colon. The ascending and descending colon was not enhanced which indicated ischaemia, but there was normal enhancement of small bowel (see figure 1A–D) and no biliary tract obstruction nor pancreatitis. Free fluid was noted in the pelvis, but otherwise normal appearances of kidneys, spleen and adrenal glands.
Figure 1.
(A) Coronal plane view of CT abdomen and pelvis demonstrating non-enhancement of the caecum and ascending colon and thickening associated with colitis. (B) Sagittal plane view of CT abdomen and pelvis demonstrating non-enhancement of the caecum and ascending colon. (C) Axial plane view of contrast CT abdomen and pelvis demonstrating moderate small bowel wall thickening and oedema. (D) Axial plane view of constrast CT abdomen and pelvis demonstrating significantly thickened large bowel in keeping with colitis.
Stool culture panel including Clostridium difficile, Shiga toxin producing-E. coli (particularly E. coli 0157:H7) and Giardia lamblia was importantly negative.
Histology of the large bowel revealed superficial mucosal haemorrhage and circumferential lymphocytic phlebitis affecting all layers of bowel wall associated with numerous thrombi. Arterial thrombi were also present as well as intraluminal fibrin and microthrombi in small vessels (see figure 2A–C). There was no evidence of malignancy and a conclusion of severe changes of ischaemic colitis due to a primary vascular disease was given. Peritoneal fluid analysis revealed blood constituents and reactive mesothelial cells, suggesting inflammation. Overall, this indicated that the disease process was likely to be of haematological origin rather than a surgical pathology.
Figure 2.
(A) Medium power magnification showing damaged mucosa with disintergrating colonic crypts, loss of the surface epithelium and inflammation spilling up through the ulcerated surface. (B) Higher power magnification of (A) showing the same features. These are typical of ischaemic colitis. (C) High magnification and has two small vessels with inflammatory cells infiltrating their walls.
Postsurgery, the patient remained in a medically induced coma and experienced further drop in haemoglobin (151–64 g/L in 2 days) and severe hypoalbuminaemia (27–11 g/L), persistent thrombocytopaenia (platelets 27 000 cells/mm3) and she required multiple transfusions.
After consulting haematology, several further investigations were requested. Serial peripheral blood films revealed numerous schistocytes, spherocytes and neutrophils indicating TMA and an acute inflammatory process (see figure 3A–D). A rheumatology screen including antinuclear antibody, rheumatoid factor and antitissue transglutaminase antibody was negative. A direct Coombs test for autoimmune haemolytic anaemia was also negative. Importantly, serum haptoglobin level was found to be 0.12 g/L, indicating haemolytic anaemia. Complement was found to be low (C3 0.56 g/L and C4 <0.05 g/L) as well as raised immunoglobulins (IgG 3.20 g/L, reference range 6.0–16.0; IgA 0.42 g/L, 0.8–3.0; IgM 0.24 g/L, 0.4–2.5). A hepatitis and HIV antibody screen was negative. The patient was then transferred to a specialist hospital. An ADAMTS13 assay (A Disintegrin And Metalloproteinase with Thrombospondin type 1 motif, 13, also known as von Willebrand factor-cleaving protease), distinctly associated with TTP, was also normal (54.00%).
Figure 3.
(A) Peripheral blood film (day 2) demonstrating numerous schistocytes (in keeping with TMA), neutrophils and spherocytes (associated with haemolytic anaemia). (B) Peripheral blood film (day 3) demonstrating numerous schistocytes and spherocytes, with decreased number of platelets. (C) Peripheral blood film (day 4) demonstrating schistocytes, neutrophils and further decrease in platelets. (D) Peripheral blood film (day 5) demonstrating numerous spiculated red blood cells (echinocytes), which are associated with dehydration and renal disease, as well as further schistocytes and minimal platelets. TMA, thrombotic microangiopathy.
A subsequent peripheral blood film revealed microangiopathic haemolytic anaemia and a presumed diagnosis of non-specific TMA was made. A molecular analysis investigating JAK2 617F mutation (associated with polycythemia vera) was also negative.
After the patient showed signs of neurological deterioration (day 15), she had an MRI head scan which revealed that she had suffered several small ischaemic infarctions affecting the right cerebral hemisphere (figure 4). The patient then underwent definitive plasmapheresis treatment. Complement levels (C3/C4) normalised after one exchange. Serum haptoglobin levels performed after this were found to be 1.51 g/L, importantly signifying a resolution to this event.
Figure 4.
MRI head diffusion-weighted image (day 8)—demonstrating multiple tiny foci of restricted diffusion (microthrombi) throughout the frontal and parietal lobes bilaterally, likely related to vasculitis due to microangiopathy. These findings concur with the patient’s focal neurological deficit.
Differential diagnosis
On initial presentation, the clinical features including fever, hypotension and tachycardia as well as abdominal pain initially suggested intra-abdominal sepsis and the patient was treated with intravenous antibiotics and supportive measures. Once ischaemic colitis and free fluid in the pelvis were identified on CT scan, this opened up possibilities including malignancy, volvulus or thrombosis of an artery supplying the colon.
After surgical resection, the subsequent haematological events, histology and blood film indicated a type of TMA. Typical and atypical Haemolytic Uraemic Syndrome (HUS) as well as TTP were postulated; however, the presenting symptoms and investigations for these ruled them all out. Typical HUS demonstrates the prodrome of bloody diarrhoea, abdominal pain and vomiting that were observed in our patient’s case, however, usually presents over several days and is preceded by infection (typically Shiga toxin producing E. c oli or Streptococcus pneumoniae), which were not present. The condition is usually less severe and often resolves with supportive therapy. Typical HUS also typically occurs in children and is less common in adults.2
There was also a strong suspicion of atypical HUS due to the initial thrombocytopaenia, low complement levels, acute kidney injury and neurological dysfunction; however, this was negated due to rapid complete recovery of renal function and absence of uncharacteristic bloody diarrhoea. This was also further confirmed by normal complement genetic results.
A normal ADAMTS13 test, absence of profound neurological symptoms and lack of purpura ruled out TTP.
Other possibilities were considered and a bone marrow biopsy showed no evidence of a B cell clonal disorder or cryoglobulin.
The patient was treated for a non-specific TMA, as there was evidence of Coombs-negative haemolytic anaemia, thrombocytopenia and ischaemic end-organ damage.
Treatment
On admission, the patient required aggressive intravenous fluid resuscitation and stabilisation. She was treated with broad spectrum intravenous antibiotics (amoxicillin, gentamicin and metronidazole). Once stable and after appropriate investigations, the patient was consented for and underwent a laparotomy. The surgeon found the entirety of the large bowel to be discoloured with a green-grey tone, pulseless and aperistaltic. There were features of fulminant pancolitis. A total colectomy and Brooke ileostomy were performed. There were no intraoperative complications and the patient was also transfused two units of red blood cells during the procedure.
The patient remained on the surgical high-dependency unit in a medically induced coma after the operation. Due to worsening haemolytic anaemia and hypoalbuminaemia, the patient was transfused altogether nine units of human albumin solution and 49 units of octaplex (prothrombin complex) in order to prevent further bleeding and maintain oncotic pressure in the postoperative period. The patient specifically was not transfused any platelet products, as risks far outweighed benefit as the patient was suffering a thrombotic event. She also had a further three units of red blood cells due to microangiopathic haemolytic anaemia. The patient remained stable while further investigations were carried out. The use of the monoclonal antibody eculizumab was considered, but disregarded once Atypical Haemolytic Uraemic Syndrome (aHUS) was ruled out.
Once a non-specific TMA was identified as the most probable diagnosis, the patient was promptly transferred to a major hospital for further investigation. Unfortunately, the patient then had a thrombotic crisis and suffered several small cerebral ischaemic infarcts, secondary to microangiopathy. This was managed conservatively. The clinical decision was made for the patient to undergo 5 days total of plasmapheresis and intravenous methylprednisolone. This astonishingly resulted in rapid, successful resolution of the TMA event. The patient was subsequently brought out of her medically induced coma, but unfortunately due to the cerebral events, she had bilateral leg weakness, left-sided hearing impairment and was unable to walk independently for several weeks. She received extensive physiotherapy and did not suffer any further ischaemic or thrombotic events for the remainder of her inpatient stay. Once the patient was medically fit for discharge, she was commenced on warfarin anticoagulation as prophylaxis to prevent further future thrombotic events as a result of a TMA of unknown aetiology.
Outcome and follow-up
After extensive rehabilitation, the patient was able to walk again and was discharged from hospital 2 months after admission. She had a further 6 months off work to recover and attended several follow-up appointments with haematology and general surgery.
After discharge from hospital, an outpatient nerve conduction study and Electromyography (EMG) were performed to investigate ongoing bilateral foot drop and revealed a possible insult to the lumbar plexus. An urgent MRI lumbar spine scan revealed no nerve injury nor any nerve compression. Repeat paraprotein screen and complement screen were both stable. Vascular endothelial growth factor level performed to rule out POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, monoclonal protein, skin changes) was raised; however, the result was of indeterminable significance as she had widespread TMA at the time of testing. Complement genetic testing was negative, further refuting aHUS. Further haematological investigations have failed to identify a specific condition thus far and investigations are ongoing, we intend to follow-up further developments and update the case report accordingly. In addition, due to the unknown aetiology, the cause has not been completely managed or treated, therefore a risk of further future TMA still remains.
The patient developed an adverse reaction to warfarin and it had been changed to sinthrome (acenocoumarol). She will remain on anticoagulation indefinitely. A diagnosis of non-specific TMA remains. Her haematology consultant justified lifelong anticoagulation on the basis that the patient suffered several minor cerebral infarctions causing lasting neurological deficit and stated that a further thrombotic event (as a result of a possible future TMA) could be life-threatening, therefore anticoagulating her would reduce this risk until such time that a conclusive aetiology is possibly found.
On follow-up in general surgery outpatient clinic, the patient stated that she has no desire for any further surgery, including ileostomy reversal, therefore was discharged from surgical follow-up and has been managing her stoma well in the community. She has recommenced work through phased return and continues to enjoy her hobbies of hiking and travelling.
Discussion
TMA occurs when an insult causes damage to the endothelium of small blood vessels which causes release of coagulation factors, leading to platelet aggregation and the formation of thrombi. These narrow and occlude the microvasculature, causing ischaemic end-organ damage—in our case this affected the colon.3
TMA can typically be characterised by the triad of Coombs-negative haemolytic anaemia with histological evidence of schistocytes in blood, thrombocytopenia (microangiopathic haemolytic anaemia) and ischaemic end-organ damage.2 They are a rare, yet life-threatening and unpredictable group of haematological conditions, which highlights the clinical urgency of early diagnosis.
There is no specific initial treatment for TMA and instead supportive therapies and rapid elimination of the underlying cause are recommended.4 A comprehensive evidence-based pathway for TMAs has been suggested by a specialist group of Mayo Clinic physicians, highlighting several differential diagnoses, investigations and treatment pathways.5 They recognise that the diagnostic workup is often complex with a broad range of possible differential diagnoses, and that uncertainties may lead to delayed diagnosis and discrepancies between management plans.5
Our patient presented with severe ischaemic colitis for which literature currently advocates surgical exploration when there are features of acute ischaemia and peritoneal signs, even if the aetiology is unclear.6 This guideline was followed by the consultant surgeon and the ischaemic bowel was resected, with the aetiology only later being revealed to be TMA after numerous investigations. Following surgery, the supportive management provided (numerous transfusions, intravenous fluids and intensive) kept the patient stable; however, she unfortunately progressed to have further thrombotic events in the form of cerebral infarction. Plasma exchange was commenced in the absence of the recommended positive ADAMTS13 (TTP) and abnormal complement genetics and assay (HUS) and resulted in an excellent outcome. This suggests scope for further application of plasma exchange in TMA.
Plasma exchange is regarded as a popular and proven therapy for treating primary TMA.7 It is the mainstay of treatment in acute TTP reducing mortality rates from 90% to 20%,8 and Fox et al suggest that it should be started within 4–8 hours of presentation in any case of possible TTP until it has been excluded with an ADAMTS13 assay.3 In aHUS, a combination of eculizumab and plasma exchange have shown successful remission in several reports.9 10 However, a conflicting review by Kim et al indicates that it does not have any therapeutic benefit in several other TMA syndromes, including malignancy-associated TMA, where the focus is cessation of the causative factors and supportive therapy.11
A similar case report involved a patient who had developed aHUS.9 A 50-year-old woman presented with bloody diarrhoea, abdominal pain and developed pancolitis, acute renal failure and profound thrombocytopaenia. She underwent total colectomy and stoma formation. Typical HUS and TTP were ruled out through further investigation. The patient’s renal failure and neurological status significantly worsened, requiring intubation and haemodialysis. MRI head revealed a left-sided infarction. The patient was anticoagulated and treated with eculizumab, making a remarkable recovery. Despite a similar course and prodrome, our case of TMA was differentiated from aHUS due to the clear schistocytosis and recovery of renal function not observed in literature on aHUS.4 11 Comparison of these cases demonstrates the stark similarities and overlap between different TMA’s and the challenge posed to clinicians in formulating an accurate diagnosis and initiating appropriate interventions.
In conclusion, this case report highlights a TMA presenting with end-organ ischaemic damage manifesting as ischaemic colitis and cerebral infarction. Prompt and accurate diagnosis is complicated by the differing clinical signs and symptoms associated with each condition, delaying critical interventions and supportive therapies. TMA continues to remain a challenging, complicated group of conditions; however, further research and new guidelines are beginning to shape how clinicians differentiate and manage them.
Patient’s perspective.
I was shocked at how quickly it all came on and progressed! It was very hard to believe. On the day that the symptoms started, I had gone to work teaching swimming and very quickly felt nauseous and I had a fever. I then began suffering hot and cold chills, vomiting and I passed several loose motions with blood in a short space of time. My partner phoned an ambulance and I was taken to hospital. I could not really remember much as I felt very unwell and confused. I had been put in a medically induced coma for a few weeks. After being brought out of the coma I could not walk, having suffered mini strokes and I feared that I would not be able to walk again; however, after months of physio I made good progress. I still have some bilateral foot-drop, ongoing left leg pain and I suffer symptoms with my left ear. I feel I have otherwise made a good recovery considering the severity of the condition. I have been managing well with the stoma, and I have decided that I would not like further surgery to reanastomose the bowel. I have returned to work after 6 months off and I have been receiving further follow-up in haematology and surgical outpatients department. My diagnosis still remains unclear after several investigations and I am still hoping to find out more.
Learning points.
Thrombotic microangiopathies (TMAs) are a collection of rare life-threatening conditions, which can manifest unpredictably and ultimately result in organ ischaemia, requiring urgent supportive treatment. It is important to have a low threshold for suspicion as early investigation and intervention can be life-saving.
The varied presentation involving multiple organ systems requires cooperation between a multidisciplinary team in order to provide coordinated and comprehensive care.
There is a growing body of evidence supporting the use of plasma exchange in primary TMA as a potentially life-saving intervention.
Recommendations for the diagnosis and management of TMA are available through the British Society for Haematology; however, there are currently no national guidelines available.
The role of surgery in the management of TMA with gastrointestinal manifestations.
Acknowledgments
I would like to thank Dr JY Soh (Honorary Clinical Lecturer—University of Birmingham) for his contributions towards data collection, Dr Kiran Randhawa (Consultant Radiologist, Russells Hall Hospital) for providing the radiology images, and the histopathology department at Russells Hall Hospital for providing histology slides and reporting. I would finally like to thank all the staff involved in the care of the patient.
Footnotes
Contributors: IN acts as the corresponding author and was involved in the initial planning of the case report, request for permissions and consent, collection and interpretation of data and follow-up relating to haematology and medicine at the patient’s initial hospital, drafting and revising versions of the work and approval of the final version as well as agreeing to be accountable for all aspects of the work. He is also responsible for responding to emails and acting upon if reports are unsubmitted. He was also involved in assisting in collection of further data to make it acceptable for resubmission. LMIS was the first author on this occasion and had a significant involvement in the initial planning of the case report, collection, analysis and interpretation of data and follow-up relating to surgery, the research of similar published articles and review, drafting and revising large sections of the work and coapproval of the final version as well as agreeing to be accountable for all aspects of the work. She has also footed the majority of the work involved in the resubmissions, contacting relevant and involved clinicians and redrafting of the work to make it acceptable for resubmission. KDS contributed to the conceptualisation, design and layout of the work, as well as providing data collection for the surgical aspect of the case and drafting sections of the work including investigations and treatment. She was also involved in approval of the final version of the work as well as agreeing to be accountable for all aspects of the work. AADS put forward the concept of this work being submitted, facilitated planning and acquiring the relevant permissions. He also assisted in data collection, interpretation of surgical findings and accessing of reports and images. He was also involved in revision of the work for intellectual content and was mainly involved in approving the final draft. He also agrees to be accountable for all aspects of the work.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Patient consent for publication: Obtained.
Provenance and peer review: Not commissioned; externally peer reviewed.
References
- 1. Scully M, Hunt BJ, Benjamin S, et al. Guidelines on the diagnosis and management of thrombotic thrombocytopenic purpura and other thrombotic microangiopathies. Br J Haematol 2012;158:323–35. 10.1111/j.1365-2141.2012.09167.x [DOI] [PubMed] [Google Scholar]
- 2. George JN, Nester CM. Syndromes of thrombotic microangiopathy. N Engl J Med 2014;371:654–66. 10.1056/NEJMra1312353 [DOI] [PubMed] [Google Scholar]
- 3. Fox LC, Cohney SJ, Kausman JY, et al. Consensus opinion on diagnosis and management of thrombotic microangiopathy in Australia and New Zealand. Nephrology 2018;23:507–17. 10.1111/nep.13234 [DOI] [PubMed] [Google Scholar]
- 4. Masias C, Vasu S, Cataland SR. None of the above: thrombotic microangiopathy beyond TTP and HUS. Blood 2017;129:2857–63. 10.1182/blood-2016-11-743104 [DOI] [PubMed] [Google Scholar]
- 5. Go RS, Winters JL, Leung N, et al. Thrombotic microangiopathy care pathway: a consensus statement for the Mayo clinic complement alternative Pathway-Thrombotic microangiopathy (CAP-TMA) Disease-Oriented group. Mayo Clin Proc 2016;91:1189–211. 10.1016/j.mayocp.2016.05.015 [DOI] [PubMed] [Google Scholar]
- 6. Sreenarasimhaiah J. Diagnosis and management of intestinal ischaemic disorders. BMJ 2003;326:1372–6. 10.1136/bmj.326.7403.1372 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Nguyen TC, Han YY. Plasma exchange therapy for thrombotic microangiopathies. Organogenesis 2011;7:28–31. 10.4161/org.7.1.14027 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pereira A, Mazzara R, Monteagudo J, et al. Thrombotic thrombocytopenic purpura/hemolytic uremic syndrome: a multivariate analysis of factors predicting the response to plasma exchange. Ann Hematol 1995;70:319–23. 10.1007/BF01696619 [DOI] [PubMed] [Google Scholar]
- 9. Ohanian M, Cable C, Halka K. Eculizumab safely reverses neurologic impairment and eliminates need for dialysis in severe atypical hemolytic uremic syndrome. Clin Pharmacol 2011;3:5–12. 10.2147/CPAA.S17904 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Cavero T, Rabasco C, López A, et al. Eculizumab in secondary atypical haemolytic uraemic syndrome. Nephrol Dial Transplant 2017;32:466–74. 10.1093/ndt/gfw453 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Kim JJ, Goodship THJ, Tizard J, et al. Plasma therapy for atypical haemolytic uraemic syndrome associated with heterozygous factor H mutations. Pediatr Nephrol 2011;26:2073–6. 10.1007/s00467-011-1944-4 [DOI] [PubMed] [Google Scholar]