The literature on psychiatric illness in surgical patients is limited and mostly confined to small patient samples within surgical subspecialties. The findings suggest that depression, anxiety, and alcohol use disorders are more prevalent, both pre and postoperatively, than in the general population.¹ In addition to pre-existing psychiatric illness, surgery may be further complicated postoperatively by delirium, acute stress disorder, posttraumatic stress disorder, new onset mood disorders, or substance withdrawal.

Studies suggest that common mood and stress-related disorders may be overlooked in the immediate postsurgical phase.² Rates of depression in the immediate postoperative period range between 10% and 14%, with rates in trauma patients nearing 40% within 1 to 3 months of surgery.³ Untreated, postsurgical depression impacts subjective quality of life scores that may persist several years following surgery and is commonly associated with poorer surgical outcomes.

Psychiatry plays an important role in the care of the surgical patient. This chapter reviews the role of the psychiatrist in evaluating patients preoperatively, and in managing depression and associated psychiatric conditions during the postoperative period. Many surgical subspecialties carry unique risks for the development of depression and these are reviewed in greater detail.

The preoperative period is a time when information can be collected about a patient’s current and past psychiatric history. For certain types of surgeries such as solid-organ transplantation and bariatric surgery, preoperative psychiatric evaluation is routinely recommended in order to identify psychosocial issues that may affect the postoperative course and outcome. Early identification of depression or other psychiatric conditions allows for adequate treatment. Decisions can also be made about maintaining psychiatric medications prior to surgery, ensuring that psychiatric illnesses are monitored postoperatively, and assessing the adequacy of social support postoperatively.

Antidepressant use is prevalent in patients undergoing elective surgery. As many as 35% of patients undergoing elective surgeries take antidepressant medications.⁴ It used to be recommended that patients discontinue antidepressant medications prior to surgery. The selective serotonin reuptake inhibitors (SSRIs) and selective serotonin and norepinephrine reuptake inhibitors (SNRIs) impact platelet function, which could affect risk of bleeding. Tricyclic antidepressants have been associated with electrocardiographic changes and vulnerability to arrhythmias during anesthesia. Monoamine oxidase inhibitors raise concern about the risk for adverse effects if combined with sympathomimetic agents, opioids or serotonergic medications during surgery.

For patients with severe depression, however, discontinuation of antidepressant medications may place them at increased risk for recurrence of their depressive symptoms or for discontinuation syndromes. And recent studies suggest that antidepressants may be safely continued prior to surgery. When the use of SSRIs and SNRIs has been examined in cardiac patients, they have not been shown to increase the risk of bleeding in the perioperative period following CABG surgery.⁵ Likewise, patients who continued to receive tricyclic and tetracyclic antidepressants prior to surgery had no increased incidence of arrhythmia or hypotension, and had a decreased risk of depressive symptoms or delirium postoperatively.⁶ In addition, a study of monoamine oxidase inhibitor use prior to surgery found no increase in the incidence of tachycardia or hypertension during the operative course. In this study, many patients received sympathomimetic drugs during anesthesia and were treated with opioids postoperatively. Only one patient (receiving meperidine) had an adverse reaction.⁷ Whenever there is concern about the advisability of continuing a psychiatric medication, particularly with monoamine oxidase inhibitors, collaboration between psychiatry, anesthesia, and surgery is indicated.

The immediate postoperative period presents several challenges for the assessment and treatment of depression. Delirium can mimic many psychiatric illnesses, and the hypoactive form of delirium is often misdiagnosed as a primary depressive disorder. Anesthesia, medication changes, and pain all contribute to the complexity of medical management during this period. Adding to the difficulty, patients may be unable to report their symptoms consistently or take medications by mouth for hours to weeks following surgery. Patients with depression report more pain, and conversely, inadequate pain control increases the risk of developing postoperative depression.

Delirium, an acute change in mental status marked by impairment of attention, remains a common and serious complication among hospitalized surgical patients. The syndrome of delirium can involve hallucinations, agitation, sleep disturbance, affective changes, and disruption of cognition, causing risk of injury as well as subjective distress.

The prevalence of delirium in the hospital ranges between 14% and 56%, and at least doubles the likelihood of mortality over the subsequent year.^8,9,10 Delirium has also been found to be an independent risk factor for the subsequent development of dementia.¹¹ One particularly vulnerable period for developing delirium is in the days following surgery, when rates as high as 70% have been reported.¹² Reviews of delirium in patients following orthopedic or cardiac surgeries have found widely varying incidence rates from 3.6% to 53.3% and 13.5% to 41.7%, respectively.^13,14

The hyperactive type of delirium, marked by agitation and dyssomnia, receives more clinical attention, often requiring medication and restraints. Yet, the hypoactive type where patients appear withdrawn, less interactive, and “quietly confused” is often undiagnosed or misdiagnosed as depression. Hypoactive delirium lacks the physical agitation often targeted by antipsychotic medication.¹⁵ In the ICU setting following cardiac surgery, hypoactive delirium is more common than hyperactive delirium and is an independent risk factor for prolonged mechanical ventilation.¹⁶ Given its relatively high prevalence and morbidity during this period, postoperative changes in mood should be considered delirium until proven otherwise. While clinical presentations can vary widely, Table 22-1 contrasts the general features of delirium and depression.

Typically patients with hypoactive delirium deny depressed mood, but may endorse associated symptoms, such as low energy, anxiety, difficulty concentrating, and sleep disturbance. Performing a cognitive exam, with special consideration to attentional tasks (e.g., saying the months of the year backward, or reciting digits backward), as well as direct inquiry into the patient’s mood are helpful in distinguishing hypoactive delirium from depression. Delirium is more typically associated with visual hallucinations, worsened symptoms at night (“sun-downing”), and acute changes in cognitive impairment.

Difficulties in cognition seen in severe cases of depression, occasionally (and misleadingly) referred to as “pseudo-dementia,” generally reflect lack of effort or apathy, as opposed to true cognitive dysfunction. Auditory hallucinations, which can occur in severe cases of either condition, are still relatively more likely to occur in delirium. In some cases, especially in patients with a significant psychiatric history, or a prolonged and complicated hospital stay, distinguishing the two can be difficult. In relatively rare instances, delirious patients can appear quite tearful and endorse several signs of depression, including suicidal thoughts. In unclear situations, an electroencephalogram (EEG) can be helpful by revealing the diffuse slowing classically seen in delirium, as well in ruling out seizure activity. In a depressed individual without delirium or other neurologic disease, the EEG would be expected to be normal.

It is worth noting that delirium and depression can coexist, though the immediate objectives following surgery should be to treat the underlying causes of delirium, while managing agitation, sleep disturbance, and subjective distress with antipsychotics. Adding antidepressant therapies during this period may actually worsen confusion and is best considered when the delirium has resolved.

The goals of delirium care have moved toward earlier recognition, identifying individuals who are at risk for delirium and screening high risk patients for it, to allow for improved management. Screening tools, including the Confusion Assessment method (CAM) or Confusion Assessment Method ICU (CAM-ICU), are validated measures designed to help nonpsychiatrically trained clinicians identify and monitor cases of delirium.^17,18 Evidence based treatment guidelines, such as those formed by the UK National Institute for Health and Care Excellence, may be helpful in the prevention, diagnosis and management of delirium.¹⁹

Most antidepressants require oral administration and absorption. For many patients taking these agents prior to surgery, they are withheld at the time of the procedure, and restarted within a few hours to days. In some situations, however, oral antidepressants may need to be withheld for longer periods. Many antidepressant pills can be crushed and administered through a nasogastric tube, a common approach for patients in surgical intensive care. However, extended release formulations, such as those found with bupropion, venlafaxine, paroxetine, and duloxetine, generally cannot be administered this way. Absorption may be improved with alternative formulations, such as orally disintegrating tablets (mirtazapine) or liquids (citalopram, sertraline, fluoxetine, lithium).

Unfortunately, there are very few parenteral alternatives. Intravenous citalopram, comparable in efficacy and tolerability to the oral version, has been produced but is not routinely available in most hospitals.²⁰ Transdermal selegiline is approved for treatment of major depression, but as a monamine oxidase inhibitor its use in the hospital is limited by potential drug interactions and risk of serotonin syndrome.²¹

In the confused postsurgical patient, withholding antidepressants may be prudent in some cases. Agents with significant anticholinergic activity, such as amitriptyline, imipramine, nortripyline, doxepin, and paroxetine may directly contribute to the incidence and duration of delirium.²² Cases of delirium have been reported with bupropion, presumably due to its relatively unique inhibition of dopamine reuptake.²³

Drug–drug interactions also deserve special consideration. For example, co-administration of the antibiotic linezolid, itself a monamine oxidase inhibitor, with serotonergic agents is associated with an elevated risk of serotonin syndrome.²⁴ Serotonin toxicity can also arise from concomitant use of SSRIs and tramadol for pain.²⁵ Fluoxetine and other psychotropic medications have been associated with increased INR and bleeding risk in patients taking warfarin.²⁶

When antidepressants need to be discontinued, it is ideal to reduce them gradually, though this is rarely possible in the immediate postoperative period. Nonetheless, every effort should be made to taper the medication, even if briefly, to avoid discontinuation reactions. This is especially important in instances where the half-life of the agent is short and the baseline dose is relatively high. Notably, paroxetine, venlafaxine, desvenlafaxine, duloxetine, all have relatively short half-lives (while that for fluoxetine is 4–6 days). Abrupt discontinuation of these medications rarely causes dangerous adverse reactions, though headache, gastrointestinal symptoms, dysphoria, irritability, dizziness, agitation, and confusion have all been reported.²⁷ All patients in this situation will require ongoing monitoring for recurrence or worsening of their depressive symptoms, possibly after discharge.

Suboptimal pain control in the peri- and postoperative periods is an independent risk factor for development of postsurgical depression. Recent orthopedic literature highlights the association between pain and depression. One study revealed that preoperative pain predicts postoperative depressive symptoms, and postoperative depressive symptoms predict pain at discharge. This suggests that perioperative pain therapy may require treatment of both pain and depressive symptoms to achieve sufficient analgesia.²⁸ Other studies have also shown that preoperative depression and anxiety are associated with worse postoperative pain, and poorer functioning and health-related quality of life.^29,30

Sleep disorders are frequently co-morbid with depression and anxiety. Postsurgical shoulder pain, for example, is closely correlated with depression, anxiety, and sleep disturbances.³⁰ Treatment of sleep disorders improves depression scores, and vice-versa.³¹ Effective pre- and postoperative screening for mood and sleep disorders may lead to earlier recognition and treatment and therefore to improved outcomes. Consultation liaison psychiatric services may be helpful for evaluation of commonly occurring psychiatric illnesses in the surgical patient, and may also be used to assess the need for, and provide access to, ambulatory psychiatric services following the acute surgical recovery phase.

BARIATRIC SURGERY

A bi-directional relationship between obesity and depression has been described, though the mediating factors are still being investigated.³² Population-based studies suggest the odds ratio of obesity leading to depression lies between 1.0 and 2.0.³³ Bariatric surgery—now encompassing the Roux-en-Y gastric bypass, gastric sleeve, and other procedures—continues to gain popularity as an effective treatment for morbid obesity (currently defined BMI ≥ 40 kg/m²). Psychiatric screening of bariatric surgery candidates is recommended and commonplace. As a consequence, relatively high levels of psychiatric co-morbidity have been detected in this population, including mood, personality, and eating disorders.³⁴ High rates of suicidal thoughts and attempts have also been reported.³⁵ Although many factors affect surgical outcomes, greater preoperative depression is associated with less postoperative weight loss.³⁶ Although many patients experience improvement in their mood and quality of life following the procedure,³⁷ antidepressant use typically remains the same before and after bariatric surgery.³⁸ In procedures that affect absorption, such as the Roux-en-Y, medications doses may need to be adjusted, as has been suggested in a pharmacokinetic study of sertraline, which revealed significantly lower plasma concentrations and an approximate halving of the mean concentration–time curve in postbariatric surgery patients compared to controls.³⁹ Liquid formulations of antidepressants may be a useful option in situations where absorption is otherwise altered, though there have been no studies demonstrating their superiority to tablet or capsule formulations in this population.

OSTOMIES

The psychological impact of a stoma is often underestimated. Recovery from ostomy creation has been associated with clinically significant psychological symptoms—such as depression, anxiety, and embarrassment—in approximately one quarter of cases.⁴⁰ These distressing emotions generally improve over time. Better mood adjustment is associated with stoma acceptance, better interpersonal relationships, and stoma care self-efficacy.⁴¹ Men may be more likely to report depression and lower quality of life⁴² though this is controversial since another study did not find this gender difference.⁴¹

Physical symptoms, such as pain and nausea, affect quality of life, as do social factors such as limitations on travel and interference with intimacy. Routine psychosocial assessment and support, especially in the early weeks following surgery, appears to be beneficial, though there remains a need for further prospective studies on depression incidence and management in this population (Box 22-1). For patients with apparent social isolation or more chronic difficulty accepting their stoma, regular monitoring for depressive symptoms and evaluation for antidepressant therapy are recommended.

GENERAL PRINCIPLES

The psychiatric care of the patient undergoing organ transplantation takes place in three distinct phases: evaluation (and/or listing), surgery, and post-transplant. In the evaluation phase, some combination of psychiatric, psychological, and/or social assessments are common if not required at many centers. This phase includes a careful screening for psychiatric disorder, as many patients have not had formal assessment previously.⁴³ At this point, the features of depression typically resemble those of patients with end-stage organ disease, though the testing, treatment planning, uncertainty, and risks associated with the transplantation process can worsen the mood of many candidates. For some patients on a waiting list for transplant, a sense of helplessness and hopelessness becomes more pervasive as the disease progresses. Transplant support groups and close monitoring of mood are especially helpful during this period.

Following surgery, complications or deviations from patient expectations can directly impact mood, especially when the recovery is prolonged. Pain, nausea, and fatigue are common complaints even in the absence of complications. Steroids can precipitate depression and affective lability in a subset of patients, especially at the relatively high doses required following transplant. Delirium, sleep disturbance, other drug side effects (such as tremor from the commonly used immunosuppressive agent tacrolimus), and the stress on social supports can all contribute to acute depression.

In the post-transplant phase, depression may arise from guilt when the graft came from a deceased donor, as well as from excessive worry that the transplant may fail. There are also psychiatric consequences from chronic immunosuppression. Post-transplant lymphoproliferative disorder arises from B-cell proliferation following infection with the Epstein–Barr Virus (EBV) and long-term immunosuppression. Most common in the first year following transplant, rare cases can occur several years later. Heralding symptoms resemble those in primary EBV infection, including lymphadenopathy, fever, and malaise. The latter may resemble depression, a presentation potentially confounded by cognitive deficits. Generally, treatment involves a reduction in immunosuppression, as well as antivirals and chemotherapy, depending on the degree of dissemination.

Progressive multifocal leukoencephalopathy (PML), a demyelinating disease stemming from JC virus can lead to mood as well as cognitive changes. While antiretrovirals are the current treatment of choice, two small case series have found some benefit with mirtazpine in preventing the progression of PML in patients with HIV, raising an interesting question of whether antidepressants confer some neuroprotective effect in this process.^44,45

In the immediate postoperative period, distinguishing among delirium, reactive depression, and the direct effects of steroids can be difficult. Generally speaking, delirium management and supportive care are the primary goals from a psychiatric perspective during this period, with close monitoring of mood, sleep, and appetite. Antipsychotics can provide acute symptomatic improvement, especially if sleep disturbance, anxiety, or hallucinations are present. An open-label 5-week trial of olanzapine demonstrated efficacy at treating depression and mixed mood (or manic) symptoms arising from corticosteroids.⁴⁶

In spite of improvements in health status following transplant, mental distress remains relatively common, with some level of depressive symptoms present in as many as 80% of solid organ recipients in 2 years following surgery.⁴⁷ In this time period, the prevalence of more severe major depression in cardiac and lung transplant recipients has been reported as high as 28% and 32%, respectively.⁴⁸ A prospective study of renal transplant recipients found a depression prevalence of 22% immediately following surgery, and depression was significantly associated with mortality over the subsequent 5 years.⁴⁹ The association of depression with worse surgical outcomes is far from clear, however. In one study of liver and renal failure patients, a Beck Depression Inventory score in the “mild” or higher range at the time of listing was associated with a threefold decrease in risk of graft failure and mortality during the 18 months following transplant.⁵⁰

There is a small but growing literature on the psychological and ethical complications of face transplantation.⁵¹ To date, more than 20 face transplants have been performed worldwide, including 7 transplantations at Brigham and Women’s Hospital, primarily for traumatic injuries. There are few data on psychiatric outcomes. Facial disfigurement is clearly associated with major psychological morbidity, including depression and suicidal thoughts.⁵² Transplantation has the potential to improve this distress as well social functioning. One case report demonstrated a decrease in depressive symptoms 3 months after successful transplant.⁵³

In addition to the unique and devastating social impact of facial disfigurement, hospitalization for the surgeries and the recovery tend to be longer than for most solid-organ transplants. Blindness and speech difficulties are common and interfere with communication. Consequently, a thorough assessment of depressive symptoms may require considerable time and repeated visits.

Advances in surgical burn trauma care have led to a sharp rise in the number of severe burn injury survivors. These patients face unique psychological adjustments and challenges from loss of function and disfigurement. Psychiatric consultation services may be utilized in any of the commonly recognized three phases of burn care: emergent, acute (reconstitutive), and long-term adjustment phases.⁵⁴

EMERGENT PHASE

The emergent phase of burn injury is broadly defined as the time from initial onset of injury through the first 72 hours of treatment. Care during this period focuses on stabilization. Pre-existing substance abuse (percentages range from 20% to 80%) is a well-documented concern among patients with burn injury. Thus, the emergent phase of burn recovery may be complicated by alcohol withdrawal and multifactorial delirium. Burn related delirium generally occurs within the first 3 days following burn injury⁵⁴ and may present with emotional lability, tearfulness, withdrawal, apathy and other symptoms that mimic depression. The diagnosis of depression, anxiety, acute stress disorder or PTSD emerges most frequently in the later phases, and after delirium subsides.⁵⁴

ACUTE PHASE

The acute (also referred to as the reconstitutive) phase lasts from the end of acute surgical intervention to hospital discharge; or from the emergent phase to the time when the burned area is completely covered by skin grafts and/or wounds are healed. During this phase, promoting sleep, as well as expedient identification and treatment of any acute stress disorder or mood disorder, promotes faster recovery and minimizes the length of hospital stay.⁵⁴ Patients with pre-existing depression have longer hospital stays and poorer wound healing than nondepressed patients.⁵⁵ This appears to hold true as well for patients who develop depressive symptoms following burn injury. Early pharmacological interventions for burn-related depressive symptoms and optimal pain control may reduce the chances of post-burn depression, and improve functional outcomes and quality of life.⁵⁶

Allostatic load may be a potential explanation linking pain during the acute phase with development of acute stress disorder, depression and poor wound healing. Allostasis, the process of achieving homeostasis during times of stress, is mediated by the hypothalamic–pituitary–adrenal (HPA) axis. Alterations in levels of norepinephrine, epinephrine, glucocorticoids, and proinflammatory cytokines occur in response to physical stressors (such as burn trauma) as well as psychological stressors. Specifically, glucocorticoids and proinflammatory cytokines are involved in the physiological mechanisms underlying both stress and wound healing.⁵⁷ During the acute stress, this neuroendorcrine response is advantageous, for example by triggering flight from danger or improving wound healing. However, the response has negative effects if it persists chronically. Chronic elevations in glucocorticoids and proinflammatory cytokines have been linked to depression and depressed patients may have poorer wound healing even during the acute burn phase. Whether depression precedes burn injury or develops after, it exerts an effect long after the acute phase; depression is a stronger predictor of long-term quality of life in burn patients than the extent of total body surface burn area.

Long-Term Adjustment Phase