Enteropathy-associated T-cell lymphoma manifesting as non-specific abdominal pain: A case report highlighting the dangers of relying on Google Translate for clinical history taking.

We present the case of a 60-year-old man with non-specific abdominal pain. We explore how communication between doctor and patient was challenged by a language barrier. We also consider how the ability to take an accurate clinical history differed between Google Translate and an accredited medical phone translation service.

INTRODUCTION

Coeliac disease is a gluten‐sensitive enteropathy, primarily involving inflammation of the small intestine. , , , Enteropathy‐associated T‐cell lymphoma (EATL) is a rare and aggressive hematological malignancy associated with coeliac disease. , , , EATL is considered the most serious complication of coeliac disease, and it usually occurs in patients who have either on‐going exposure to dietary gluten or have refractory disease. , Refractory disease is found in patients whose enteropathy is non‐responsive to the complete elimination of dietary gluten, despite an initial period of response. EATL is most commonly manifested as recurrent diarrhea, unexplained weight loss, abdominal pain, fever, and night sweats in patients with previously well‐controlled symptoms. Treatment for EATL requires an aggressive approach, often with initial surgical resection, followed by high‐dose combination chemotherapy regimens. , , The prognosis of EATL is very poor, with 5‐year survival rates in the range of 11–20%.

As the symptoms of EATL can mimic those of coeliac disease itself, teasing out the history is challenging and requires exactly that an accurate patient history. Due to the busy and chaotic nature of the emergency department (ED) and poor access to accredited translational services, under‐pressure ED clinicians may resort to websites such as Google Translate rather than accredited translational services such as “The Big Word” when assessing non–English speaking patients. This poses a threat to accurate clinical history taking, and consequently, management and patient experience in an already disadvantaged patient group.

Patients who are not proficient in English are twice as likely to revisit the ED within 72 h of discharge. During discharge, providing safety‐netting advice is a part of routine emergency medicine practice. Suboptimal clinician communication skills, such as failure to ensure that the patient understands the diagnosis and management plan, may contribute to non–English speaking patients revisiting the ED. As repeat ED visitors with recent normal investigations may be under‐investigated on subsequent visits, the risk to non–English speaking revisiting patients is compounded.

CASE HISTORY

This case report is of a 60‐year‐old Bulgarian man who visited a level 1 ED seven times within a 20‐day period. He presented with a 2‐month history of worsening non‐specific abdominal pain. He had a past medical history of coeliac disease, including an inpatient admission in 2019 for abdominal pain that was attributed to non‐compliance with the recommended gluten‐free diet. Another relevant finding of his medical history was bilateral inguinal hernia repair. He was a non‐smoker and non‐drinker, who worked as a factory worker at a meat processing plant. He could not speak in English, and each visit to the ED required him to take two bus journeys from his home that was 40 km away.

The patient had visited the ED six times previously and had been discharged with a diagnosis of non‐specific abdominal pain based on normal blood tests and imaging results. Notably, during his fourth visit, he was referred to and reviewed by the surgical team, who arranged computed tomography (CT) imaging of the abdomen, but no acute abnormality was found. When the notes from the patient's most recent visit were reviewed, it was found that his medical history was taken using Google Translate. The patient indicated that he had symptoms of diarrhea and abdominal pain and that he had previously experienced similar symptoms. Based on the history obtained using Google Translate, the patient's symptoms indicated a chronic problem. The patient was subsequently discharged with analgesia and referred for outpatient follow‐up.

On the seventh visit to the ED, the patient was reviewed with the aid of an accredited medical telephone translation service called “The Big Word.” This service required a single room, speakerphone, and an estimated 15‐min wait to be connected with an appropriate translator. The patient could however give a detailed history in his own words. He described a 2‐month history of new, increasingly severe, spasmodic abdominal pain associated with weight loss, diarrhea, nausea, and fatigue. He was adamant that he had strictly adhered to the recommended gluten‐free diet. He had felt very well up to 2 months previously and had made no changes in his diet or lifestyle. He explained that he could not cope with the pain at home nor could he afford to continue visiting the ED because he had no access to private transport. He vomited copious amounts of undigested food in the ED. All clinical observations were within the normal range. Physical examination revealed a tender epigastrium and mild abdominal distension. The patient's routine blood investigations, including complete blood count, C‐reactive protein level, and liver function results, were all unremarkable.

Following the patient's detailed description of an acute deterioration in his health together with the examination findings and subsequent senior advice, it was decided to repeat the imaging. Abdominal radiography revealed distended bowel loops in the left upper abdomen. Although some gas was noted in the colon, the radiographic finding was reported as “suspicious for small bowel obstruction.”

Initially, the on‐call surgical team felt that a repeat surgical review would be of limited clinical benefit, given the recent normal CT and non‐definitive radiographic findings. However, after discussing the updated detailed history gained with the help of an appropriate interpretation service, the patient was reviewed and admitted for observation. Repeat CT of the abdomen revealed a jejunal lesion suspected of being a primary neoplasm.

DIFFERENTIAL DIAGNOSIS

The primary differential diagnosis was pain secondary to non‐compliance with a gluten‐free diet. Tuberculosis was included in the differential diagnosis for the jejunal lymphoid lesion.

INVESTIGATIONS

Computed tomography of the abdomen and pelvis within 24 h of re‐admission to the hospital and 15 days following previously normal CT, showed a lesion in the proximal jejunum with adjacent adenopathy, indicative of primary small bowel neoplasm (see Figure 1).

FIGURE 1

Computed tomography scan of the abdomen shows a lesion in the proximal jejunum with adjacent adenopathy, indicative of primary small bowel neoplasm

TREATMENT, OUTCOME, AND FOLLOW‐UP

The patient underwent emergency laparotomy for subacute bowel obstruction and resection of the jejunal lesion within 48 h of presentation. He did well postoperatively, and he was discharged to await the histology results and multidisciplinary team discussion. Histopathology of the lesion following resection confirmed EATL. He was subsequently referred for hematology follow‐up on a red flag basis.

The patient was reviewed by the hematology department, with the first review conducted 1 month after discharge from the surgical team. An accredited translator was present for the appointments. The hematology consultant explained the diagnosis to the patient and that combination chemotherapy treatment was advised.

DISCUSSION

Abdominal pain is one of the most common presentations of patients visiting the ED, accounting for 5–10% of all patient visits. The majority of abdominal pain presentations to ED are with non‐specific abdominal pain; where history, examination, and investigations cannot, at least initially, point to a clear organic cause. Despite the frequency of abdominal pain presentations, it can be a sign of a serious underlying disease, and delayed diagnoses can have severe adverse outcomes. When symptoms are non‐specific, risk stratification and management decisions become even more challenging, thus necessitating the need for a clear and accurate clinical history from the patient.

Although it is a rare cause of non‐specific abdominal pain, the progression of coeliac disease to EATL should be considered within the differential diagnosis in coeliac patients presenting with abdominal pain, diarrhea, weight loss, or night sweats. This is particularly prudent when the patient presents with recurrence or worsening of symptoms after a period of good symptomatic control. As EATL requires aggressive treatment and carries a guarded prognosis, it is important to reduce the diagnostic delay so that treatment can be started as soon as possible.

In the present case, the language barrier between the patient and doctor posed a communication challenge that complicated clinical history taking and subsequent diagnosis. It also contributed to the difficulty in determining whether the patient had actually been able to comply with the recommended gluten‐free diet. Repeat ED visits for a seemingly chronic problem coupled with recent normal investigations may have contributed to the hesitation in repeating CT imaging before this visit and an initial degree of difficulty in securing surgical re‐assessment.

Language barriers are an established major obstacle to accurate communication in the ED, leading to potential and actual medical errors. Reviewing patients who are not proficient in English in an ED setting can be challenging for clinicians. More importantly, it can be stressful and disheartening for the patients seeking our care. Existing research concludes that patients who are not proficient in English are more likely to be misdiagnosed and to be dissatisfied with the care provided when compared with English‐speaking patients.

The General Medical Council states that it is the duty of a doctor to “make sure that arrangements are made, wherever possible, to meet the patients' language and communication needs”. While the gold standard is an accredited in‐person translator for scheduled appointments, this is frequently not possible in the ED setting. Clinician advice regarding translation services is predominately based on research in primary care or inpatient settings, which often fails to address the communication challenges encountered specifically in the ED. Time pressure, potential distraction, sense of urgency, interruptions by multiple staff members, and lack of prior information on patients are all factors that can make communication in ED uniquely challenging. As a result, using non‐accredited translation services, such as Google Translate, can feel like the only option for under‐pressure ED clinicians. Unaccredited online translation websites such as Google Translate have been shown to be insufficiently accurate for medical history taking and should therefore be avoided if possible. Accredited medical translational services reduce the risk of communication error and re‐attendance in the ED setting. ,

PATIENT'S PERSPECTIVE

Via Bulgarian translator while undergoing inpatient chemotherapy

I was very frustrated because I knew that something was very wrong with me, but I kept being sent home from the Emergency Department with Co‐Codamol. It was very difficult, as I could not put into words what I wanted to say. I had been well, and then I was sick. I knew something was different. The last time I visited the hospital, I was begging not to be sent home.

CONFLICT OF INTEREST

None declared.

AUTHOR CONTRIBUTIONS

Dr. Deirbhile Leyden (SHO in Emergency Medicine, South Eastern Health and Care Trust Northern Ireland) dleyden01@qub.ac.uk. Dr. Niamh Melby (Consultant in Emergency Medicine, South Eastern Health and Care Trust Northern Ireland). DL prepared the manuscript. NM read and critically appraised the manuscript and advised on revisions.

CONSENT

Written consent was obtained from the patient for this publication via an NHS‐accredited translator.