Somatic symptom severity, depression and anxiety associations with pancreatitis and undifferentiated abdominal pain in surgical inpatients

Somatic Symptom Disorder is a psychiatric diagnosis that describes the experience of physical symptoms and associated distress, that is disproportionate to recognized organic pathology. Somatic symptom severity (SSS) may be associated with some surgical diagnoses; particularly the complex pain associated with pancreatitis, or the diagnostic ambiguity of undifferentiated abdominal pain (UAP). We aimed to estimate the prevalence of SSS in different diagnostic groups in surgical inpatients with abdominal pain; and to estimate the magnitude and direction of any association of SSS, anxiety and depression.


Introduction
Somatic Symptom Severity describes the experience of somatisation: physical symptoms, most commonly pain, associated with disproportionate and persistent anxiety and distress. 1Somatisation disorders have prevalence rates that vary in different clinical populations: 25% in primary care, 2 35% in medical inpatients 3 and 52% in surgical inpatients. 4Somatisation disorders are associated with increased health service use: 1.5 times as many primary care visits, 3 times as many hospitalizations, and more major outpatient procedures and Emergency Department visits than patients without somatisation. 5he symptoms of somatisation may be broad, including but not limited to: lethargy, diarrhoea, dysmenorrhea and pain, 1 often leading patients to multiple medical specialists during their lifetime.As a result, different specialties have recognized common presentations for which a unifying organic diagnosis cannot fully explain the patient's somatic symptoms and health focused distress, e.g.Irritable Bowel Syndrome (IBS) and Non-Ulcer Dyspepsia (Gastroenterology), Chronic Cough (Respiratory Medicine), Chronic Pelvic Pain Syndrome (Gynaecology) and Chronic Fatigue (Neurology). 6There may be an overlap between these specifically named functional syndromes, and the psychiatric diagnoses of somatisation.Somatising patients often complain of symptoms in multiple bodily systems, may meet diagnostic criteria for several different functional syndromes 6 and may be frequently comorbid with common psychiatric disorders, such as anxiety and depression. 7omatisation prevalence has been estimated in various secondary care populations including 91.7% of 178 patients with Chronic Pelvic Pain Syndrome, 8 and 44% in 116 adult endoscopy patients with unexplained gastro-intestinal complaints. 9espite increased awareness of functional syndromes, and their relationship to somatization diagnoses, General Surgery has not readily recognized any specific 'functional' diagnoses.However, two surgical diagnostic groups have been considered to have a somatic component to their symptomatology, namely pancreatitis and undifferentiated abdominal pain. 10This is largely due to the shared characteristics of these conditions with common somatic symptoms (described above), increased prevalence of comorbid psychiatric disorder, uncertain symptom pathophysiology and difficult doctor-patient relationships. 6cute pancreatitis is a common surgical condition, usually self-resolving with supportive care and associated with alcohol ingestion and gallstone disease. 11Chronic pancreatitis is a slow, irreversible process characterized by parenchymal fibrosis and pancreatic dysfunction, 12 and presents a diagnostic dilemma, with many papers referencing histological changes, exocrine function or invasive investigation changes such as ERCP or MRCP as diagnostic gold standards.However, none of these investigations are completely sensitive of specific, in addition they are expensive or invasive. 13For the purposes of our research, we define chronic pancreatitis as 3 or more admissions with pain and investigations consistent with pancreatitis.Pain is the most common reason for hospitalization among chronic pancreatitis sufferers, and many require 3 of more admissions during their lifetime for pain management. 14Patients with chronic pancreatitis have an increased rate of mental health diagnoses, 10 and an increased risk of suicide 15 compared to the general population, as well as alcohol and opioid substance misuse, characteristics shared with patients suffering from somatization. 16ndifferentiated symptoms refer to non-specific symptoms that have not yet manifested into an identifiable illness. 17UAP is a common inpatient diagnosis in general surgical services, for which a somatic association is often presumed. 10Patients who undergo an appendectomy for presumed appendicitis, but who fail to show histological changes consistent with appendicitis (Negative Appendicectomy -NA), are often discharged with their presenting symptoms remaining undifferentiated.Negative appendicectomy is associated with an increased risk of having 'any mental disorder' (6.69% versus 4.93%). 18e aimed to explore the association of SSS, depression and anxiety, in pancreatitis and undifferentiated pain sub-groups, in order to encourage recognition and appropriate psychological management of surgical inpatients with abdominal pain.

Aims
In a surgical abdominal pain inpatient population, (1) Describe the frequencies of SSS, depression and anxiety, stratified by surgical discharge diagnosis sub-groups.(2) Explore the association of SSS, depression and anxiety, in pancreatitis and UAP sub-groups.

Study design and population
This study was conducted at a 694-bed tertiary hospital in NSW, Australia.A cross-sectional study of consecutive adult patients admitted to the Acute General Surgical Unit (AGSU) with any form of non-traumatic abdominal pain.Full details of the study and the participant characteristics have been reported previously. 4xclusion criteria: age less than 18 years, inability to provide informed consent; or any patient admitted to other services such as general medicine.Only the index admission in the study period was used for analyses.

Ethics approval
This study was approved by the Hunter New England Human Research Ethics Committee REGIS Site Specific Agreement Reference No: 2018STE00509.

Study procedures
After admission to the surgical ward, eligible patients were approached by a member of the research team and if interested in participation were provided with an envelope containing the study questionnaires.Upon completion of the questionnaires, participants were encouraged to insert the materials into and seal the provided envelope and return it to research staff.

Measures
The PHQ-15 is a self-report instrument designed to screen for somatisation syndromes.It consists of 15 questions about the most common somatic symptoms and the degree of distress over the previous 4 weeks. 19It has been validated as a screener for DSM-5 Somatic Symptom Disorder, 20 and for SSS in medical inpatients. 21t can be used without further inquiry into the patient's medical comorbidity, as physical co-morbidity does not alter utilization significantly in primary care populations. 5he PHQ-15 is scored 0-30, with scores of 5, 10 and 15 representing cut off points for low, medium and high SSS, respectively. 22We reported PHQ-15 ≥ 10 as medium and PHQ-15 ≥ 15 as high SSS.
To assess the comorbidity of depressive and anxiety symptoms, participants also completed the PHQ-9 and GAD-7 instruments.A PHQ-9 of ≥10 points can be used to classify a DSM-IV Major Depressive Episode 23 ; with PHQ-9 scores reported as 'depression symptoms'.A GAD-7 of ≥10 points has been used as the cut-off for any DSM-IV Panic Disorder 22 ; with GAD-7 scores reported as "anxiety symptoms".

Surgical discharge diagnosis sub-groups
Discharge diagnosis was obtained from the discharge summary and then confirmed by review of operation report, histology, imaging, and blood results.An accuracy of 97.9% was found.
Discharge diagnosis was then classified into 10 sub-groups via consensus judgement by 2 surgical authors (MS and PP): • Pancreatitis • Acute and chronic • Abdominal pain • Lipase elevated 3 times the upper limit of normal • Conclusive findings on imaging.Chronic pancreatitis in this study is three or more patient reported hospital admissions for pancreatitis of any cause.Acute and chronic pancreatitis are recognized as separate conditions, but there is no widely accepted and repeated definition that we could use to separate the two.We used 3rd or more presentation as being a level of repeated exposure to the misery of the condition that might be expected to induce greater anxiety and stress due to recognition and familiarity.
UAP was classified based on the discharge summary diagnosis: UAP, based on a patients' symptoms remaining unexplained at time of discharge.

Statistical analyses
Participant characteristics were reported using descriptive statistics for age, SSS, depression, anxiety and any operation in theatre.The proportions of patients meeting the thresholds for PHQ-15 (medium and high cut-points), PHQ-9 and GAD-7 were reported for each surgical diagnostic sub-group.
The proportion of patients scoring above and below the PHQ-15, PHQ-9 and GAD-7 thresholds was calculated for patients with acute and chronic pancreatitis verses all other surgical diagnoses along with the odds ratio (OR), 95% confidence intervals (95% CI) and p-values.Similar analyses were performed for UAP verses all other surgical diagnoses.

Results
Seven hundred and thirty-one eligible admissions were invited to participate in the study and 590 questionnaires were returned (80.7%).After removal of 96 incomplete data sets and 29 repeat admissions, 465 patients were included in this study.
Of the 465 participants, 53% were female, 60% were aged 31-70 years and nearly half underwent an operative procedure (Table 1).Moderate SSS (PHQ-15 ≥ 10) occurred in over half of the participants, with a female predominance.Severe SSS (PHQ-15 ≥ 15) occurred in 20% of participants with no gender difference.Depressive and anxiety symptoms were common; 33% and 20% respectively, with no difference between genders.

Somatic symptom severity and discharge diagnosis
Apart from pancreatitis, none of the diagnostic categories differed from each other, or from the whole population, in their SSS profiles (Table 2).
Participants with a discharge diagnosis of pancreatitis did differ from the rest of the population.Twenty participants met the diagnostic criteria for acute pancreatitis (4.3%), of which 15 (75%) had a PHQ-15 score ≥ 10 and 5 (25%) had a score ≥ 15.Eighteen participants had 3 or more admissions for pancreatitis, of which 11 (61.1%) had a PHQ-15 score ≥ 10 and 6 (33%) had a score ≥ 15.Severe somatic symptom severity, depression and anxiety were all more prevalent in participants with chronic pancreatitis compared to acute pancreatitis (Table 2).
In our study we found that 68% of all participants with pancreatitis had moderate SSS, 55.3% had depressive symptoms (PHQ-9 ≥ 10) and 31.6% had anxiety symptoms (GAD-7 ≥ 10).When participants with acute and chronic pancreatitis were analysed separately a significant association between chronic pancreatitis and depression can be seen (OR = 3.47, 95% CI 1.31-9.24).
There was a significant association between moderate somatic symptoms and all pancreatitis (OR 2.11, 95% CI 1.05-4.25),with a non-significant association for severe somatic symptoms (OR 1.77, 95% CI 0.85-3.67).When acute and chronic pancreatitis were analysed separately their power was reduced, but a trend of increased moderate somatic symptom severity in participants with acute pancreatitis was seen.There was a significant association between depressive symptoms and all pancreatitis (OR 2.73, 95% CI 1.38-5.38),with a non-significant association for anxiety symptoms (OR = 1.97, 95% CI 0.97-4.03).When assessed separately a significant association between depression and chronic pancreatitis can be seen (OR = 3.47, 95% CI 1.31-9.24).There was no evidence of significant associations for the four mental health categories and undifferentiated abdominal pain (Table 3).

Discussion
In this population we have seen that moderate somatisation is present in 53%, and severe somatisation in 20%, with substantial psychiatric comorbidity across all surgical sub-groups.

Pancreatitis
It has previously been shown that patients with pancreatitis have higher rates of psychological comorbidity, with patients suffering from chronic pancreatitis having a 2-fold higher risk of developing mental disorders when compared to those with acute pancreatitis. 16Patients with pancreatitis have an increased risk of suicide 15 as well as alcohol and opioid substance misuse, characteristics that are also seen in patients suffering from somatisation. 16Alcohol, especially excessive and repeated use, is a common aetiology for pancreatitis, hence it is not surprising that there is an association between SSS and pancreatitis.As many pancreatitis sufferers have repeated admissions for pain (more than half the pancreatitic group were on their 3rd or more hospital admission) their higher scores for depressive and anxiety is perhaps to be expected.

Undifferentiated abdominal pain
In our study 16% were discharged with a diagnosis of UAP or negative appendectomy, with 54.7% having moderate and 18.8%  Note: The analysis of acute pancreatitis excludes the 18 patients with chronic pancreatitis; similarly, the analysis of chronic pancreatitis excludes the 20 patients with acute pancreatitis.
having severe SSS.The UAP group had similar SSS, depression and anxiety symptom scores, to the overall abdominal pain population.In adult patients who present to ED with UAP, less than 10% represent with pain, and less than 5% of these have their original diagnosis changed to an organic pathology. 24Most patients (88%) experience reduction or resolution of pain within 3 weeks of discharge and mortality is rare. 25plications for surgical management of comorbid surgical pathology and psychological symptoms Somatisation is associated with increased healthcare utilization, repeat hospitalisations, iatrogenic harm and decreased health related quality of life. 5It has been suggested that caution should be used when offering surgery to patients with 'poly-syndromic phenotypes' to avoid 'catastrophic consequences'. 26owever, somatising patients that also have identifiable physical pathology benefit from surgical intervention.Patients with evidence of gastro-oesophageal reflux experienced the same improvement in quality-of-life post fundoplication; whether they suffered from somatisation or not. 27Patients undergoing total hip replacement with high somatisation and depression scores experienced the same benefit from surgery as non-somatising patients. 28his study also challenges the widely held folklore held by healthcare professionals that somatisers have lower rates of 'real' pathology than non-somatisers.That there are no differences between the somatisation scores of those with a discharge diagnosis of UAP and of all other surgical diagnoses, other than pancreatitis, ought to spark a reconsideration of the likelihood of pathology in someone who displays somatising behaviours.

Strengths and limitations
This study estimated the prevalence and comorbidity of somatic symptoms by surgical diagnostic groups in an infrequently studied population.Our study used broad inclusion criteria, a large sample size and validated instruments for the key outcomes.Clinician diagnosis and treatment decisions were made blinded to study results.
The recruitment processes meant some brief admissions may have been missed and some patients who were very unwell requiring ICU, may have been excluded.The validity of using the PHQ-15 in older participants (over 60 years) is unclear.Our comparative analyses used small sample sizes in each surgical diagnostic group, resulting in relatively low statistical power.We did not adjust for any possible confounding.Nevertheless, the findings of increased psychological morbidity in pancreatitis were significant and worth further investigation in larger samples to establish more precise estimates of association.

Conclusions
In our surgical inpatient population somatic, depressive and anxiety symptom burdens were common and comorbidity with established surgical diagnosis was around 50% for moderate SSS.Surgical management in the presence of somatisation requires careful consideration of the surgical pathology and co-ordinated psychological intervention for the psychological pathology.Effective psychological interventions are available for somatic, depressive and anxiety symptoms for symptomatic and health utilization outcomes.Pancreatitis, especially those with chronic pancreatitis, had higher rates of somatic, depressive and anxiety symptom burdens than other diagnostic groups.If this result is confirmed in larger studies, pancreatitis patients may be a priority population for psychological assessment and intervention.
• UAP and histologically identified Negative Appendectomy • Appendicitis: radiologically or histologically identified • Biliary pathology: Cholecystitis, Cholangitis and Biliary Colic • Gastritis, Enteritis, Colitis and Diverticulitis • Small Bowel or Large Bowel Obstruction and Hernia • Post-operative Complication • Non-surgical pathology: i.e. gynaecological or medical diagnosis admitted under the surgical services, e.g.ruptured ovarian cyst, pneumonia or urinary tract infection.• Rectal bleeding • Other diagnosis without category, e.g.musculoskeletal pain, SMA stenosis, rectus sheath haematoma Acute Pancreatitis was classified by meeting two of three criteria as diagnostic for acute pancreatitis 11 :

Table 1
Participant demographics

Table 2
Somatic Symptom Severity, Depression and Anxiety by surgical diagnostic sub-groups

Table 3
Association between pancreatitis, undifferentiated abdominal pain and mental health symptoms