Chapter 32-1: Approach to the Patient with Unintentional Weight Loss - Case 1

Significant unintentional weight loss is defined as > 5% loss of usual body weight in the last 6–12 months and can be a harbinger of serious underlying disease. One study documented significantly increased mortality in men with unintentional weight loss compared with men whose weight was stable or increased (36% vs ≈ 15%). The evaluation of unintentional weight loss is complicated by its frequency among older adults; 15–20% of adults over 65 years of age may have unintentional weight loss, and the prevalence rises to 50–60% among nursing home residents. Body weight usually peaks about age 60 and then decreases gradually thereafter, especially after age 70, although the normal changes are small (< 1 lb/year).

It should be noted that weight loss in the elderly is often part of an overall syndrome of functional decline called frailty, which includes weakness, slowness, low level of physical activity, self-­reported exhaustion, and unintentional weight loss. When present, frailty has significant implications for morbidity and mortality in the elderly and frail patients have substantially higher morbidity and mortality that their age-matched peers.

There are a large number of diseases that can cause unintentional weight loss, which are best organized by system (see below). The most common causes of unintentional weight loss in published studies are cancer (most commonly gastrointestinal [GI] but also lung, lymphoma, and other malignancies), ≈ 19–36%; depression and alcoholism, 16%; nonmalignant GI diseases, 13%; and unknown, 22%. Endocrine disorders account for 7% of uninten­tional weight loss. Although cancer is the most common cause, it is not the cause in most patients. Mild cognitive impairment and dementia may also cause weight loss due to a combination of increased energy expenditure (due to agitation and pacing) and decreased caloric intake.

Four pivotal points are worth remembering when evaluating patients with unintentional weight loss (Figure 32-1). First, the weight loss should be documented if possible; 25–50% of patients that complain of unintentional weight loss have not in fact lost weight (and do not need to be evaluated for causes of weight loss). Elderly adults often lose muscle mass and simply look like they lost weight. Weight loss should be documented by comparing prior weights or, if these are unavailable, by finding a significant decrease in a patient’s clothing and/or belt size. It may also be helpful to see corroborating information from family and/or caretakers and trying to elicit a specific amount of weight loss from the patient.

Clinicians should verify the weight loss or document significant changes in the patient’s clothing or belt size.

The second pivotal step identifies patients with malabsorption. Patients should be asked about changes in their bowel movements, including diarrhea or large, difficult to flush or malodorous stools. Although an uncommon cause of unintentional weight loss, such symptoms suggest small bowel or pancreatic disease and direct the diagnostic search. Additionally, difficulty with defecation or changes in stool caliber raises the possibility of colorectal cancer.

The third pivotal step differentiates weight loss that is due to decreased caloric intake from weight loss that is due to increased caloric expenditure or malabsorption. Most patients lose weight due to decreased caloric intake. Weight loss associated with normal or increased intake suggests substantial catabolism or energy loss that may be seen in patients with cancer, severe chronic obstructive pulmonary disease, hyperthyroidism, malabsorption, or poorly controlled diabetes.

Fourth and most importantly: a clinician is most likely to be successful in identifying the cause of a patient’s unintentional weight loss if she identifies the “company it keeps.” Clinicians should perform a truly comprehensive history (including a review of systems, psychosocial history, and medication history) and perform a detailed head to toe physical exam (with special attention to the mental status exam) to identify any abnormalities. An initial baseline laboratory evaluation should include a complete blood count (CBC) with differential, urinalysis, renal panel, calcium, hepatic panel, fasting glucose, fecal occult blood test (FOBT), erythrocyte sedimentation rate (ESR), thyroid-stimulating hormone (TSH), HIV test, and chest radiograph. Abnormal, or potentially abnormal, findings discovered on history, physical exam, or initial labs are rarely conclusively diagnostic but often provide critical clues to the underlying diagnosis and should be thoroughly evaluated. Examples of such findings include anemia, which may be due to iron deficiency from an unsuspected carcinoma of the colon or stomach; an elevated alkaline phosphatase, which may be due to metastatic disease to the liver or bones; hematuria, which may be due to carcinoma of the kidney or bladder; or a markedly elevated ESR, which may be due to multiple myeloma, temporal arteritis, subacute bacterial endocarditis (SBE), or other chronic infection. Additionally, recommended preventive health exams should be brought up to date (eg, mammogram, Papanicolaou exam with human papillomavirus testing, colonoscopy or other appropriate colorectal cancer screening, prostate-specific antigen in selected patients and low-dose chest CT scan for smokers with ≥ 30 pack year smoking history unless they quit > 15 years previously).

Patients without any such clues may benefit from an upper endoscopy and an abdominal ultrasound. Figure 32-1 illustrates the diagnostic approach.

Interestingly, an unrevealing initial evaluation is reassuring: one study conducted in patients with weight loss showed that among those who underwent evaluation, none of those without abnormal or concerning findings on the initial evaluation were diagnosed with cancer during the study. A more recent, larger study showed that cancer may be undiagnosed antemortem but discovered on autopsy; this suggests that, while the rate of cancers diagnosed on follow-up (after the initial evaluation) is low, cancers may still be present and thus warrant ongoing clinical follow-up.

The patient’s relatively nonspecific history is typical of many patients complaining of weight loss. Patients report an unspecified amount of weight loss that is associated with anorexia. The first pivotal step in the evaluation is to verify that weight loss has, in fact, occurred.

Mrs. M’s change in clothing size suggests true and significant weight loss. The second pivotal step in evaluating patients with documented weight loss determines whether the patient has symptoms suggestive of malabsorption, such as diarrhea or floating stools.

Since the history does not suggest diarrhea or malabsorption, the third pivotal step in the evaluation of patients with unintentional weight loss differentiates weight loss due to decreased caloric intake or absorption from that due to increased caloric expenditure. Mrs. M.’s initial complaint already clarified that her weight loss was due to anorexia and decreased caloric intake. The fourth step is a comprehensive, system-based approach utilizing a thorough history, (including a past medical history, detailed psychosocial history, and review of systems), physical exam, basic laboratory tests, and completion of her age-appropriate health maintenance examinations to search for diagnostic clues.

Unfortunately, like many patients with unintentional weight loss, Mrs. M has no specific clues to suggest any particular diagnosis. Given her cachectic appearance (suggesting her weight loss is real despite the lack of documentation) and the fact that cancer is the most common cause of unintentional weight loss and a must not miss hypothesis, it is the leading hypothesis that must be evaluated at this point. Her tired appearance also raises the possibility of depression (which could be either a primary psychiatric disorder or a reaction to an underlying medical condition). Finally, nonmalignant GI disease should also be considered, since it is also a common cause of unintentional weight loss. Hyperthyroidism is also on the differential. Table 32-1 lists her differential diagnosis.

Clearly, a diagnosis is not yet apparent on history or physical exam for Mrs. M. The data suggest that when the cause of unintentional weight loss is malignant, there are usually clues on history, physical exam, or on laboratory testing. You elect to check a CBC, liver panel, renal panel, urinalysis, chest radiograph, and screening mammogram. Finally, you elect to schedule Mrs. M for a colonoscopy, since she has never undergone colon cancer screening. You discuss that this is different than usual age-appropriate colon cancer screening, for which she is above the age at which screening is still recommended; this is a diagnostic evaluation with a substantially different pretest probability.

Mrs. M’s answers to the screening questions make depression highly unlikely (LR− 0.07). Although her appearance seems antithetical to what you know about hyperthyroidism, you wonder if that possibility should be pursued.

Check the TSH on every patient evaluated for weight loss.