New tools for detecting occult monoclonal gammopathy, a cause of secondary osteoporosis

Author and Disclosure Information




Declining bone density

Standard DXA testing is used to identify patients at high risk of fragility fractures from osteoporosis. It is also the accepted way to monitor disease progression and efficacy of treatment.

However, when checking to see if a patient’s bone density has changed over time, one must recognize that variations in technique from center to center or operator to operator can produce false changes in DXA results. 3,4 The testing center should state its own level of variance (referred to as the least significant change) and should indicate whether changes in a patient’s follow-up test results are statistically significant (ie, exceed that level).

A significant decline in bone mineral density over time may indicate that the patient is either not taking his or her medications or is not taking them as directed, as often happens with oral bisphosphonates—which must be taken first thing in the morning, on an empty stomach, with only a glass of water, at least 30 minutes before breakfast, during which time the patient must remain in an upright position.5–7 But a decline also raises the suspicion of an underlying condition instead of or in addition to osteoporosis, as described in the cases above. The normal decline in bone mineral density due to aging is 0.1% to 0.2% per year. For women 5 years after menopause, the rate increases to 1% to 2% and then slows to the rate of decline due to aging. A decline in bone density to the degree seen in case 1 is more than that which could be attributed to primary osteoporosis, and so an underlying cause must be considered.

Abnormally low Z scores also raise the suspicion of secondary osteoporosis

The T score is the difference, in standard deviations, between the patient’s bone density and the mean value in a population of healthy young adults. Since bone density tends to decline with age, so does the T score.

In contrast, the Z score compares a patient’s bone density with the mean value in a population the same age and sex as the patient. When it is abnormally low, it implies greater bone loss than predicted by aging alone or greater than expected from primary disease, so a secondary disorder must be considered.8,9 This was the case in our second patient, who had a Z score of −2.4.

No specific Z score cutoff has been established. Rather, the physician should be suspicious when it is lower than about −1.0 and when something in the patient’s clinical presentation, history, or laboratory evaluation raises suspicion of an underlying condition. In other words, the Z score is useful not by itself, but in context with other information.

In a retrospective analysis of men and women with osteoporosis, Swaminathan et al9 reported that a Z score cutoff of −1.0 had a sensitivity of 87.5% for detecting an underlying cause of osteoporosis.

Again, we want to emphasize that a low Z score alone is not sufficient to make a diagnosis of a secondary cause of osteoporosis. But it is good to be suspicious when a Z score is as low as in our second case and when that suspicion is reinforced by other clinical data.


Biochemical markers of bone resorption, such as urinary NTX and the cross-linked C-telopeptide of type I collagen (CTX), have been shown to predict fracture risk independent of bone density measurements. The evidence to date supports the use of these markers in conjunction with bone density measurements to ascertain early on whether osteoporosis is responding to treatment, but their use alone to screen for osteoporosis is not encouraged.10

The markedly high level of NTX in our second patient would be unusual in primary disease—it implies a high degree of bone turnover and, in concert with the clinical information, suggests secondary osteoporosis.


If a patient has a low Z score, a declining T score, or other clues, it is critical to evaluate for causes of secondary bone loss, such as8:

  • Endocrine disorders (Cushing syndrome, hyperparathyroidism, hypogonadism)
  • Gastrointestinal disorders (malabsorption, cirrhosis, gastric bypass surgery)
  • Renal insufficiency and failure
  • Pulmonary diseases and their treatment
  • Drug use (corticosteroids, antigonadotropins, anticonvulsants, aromatase inhibitors, antirejection drugs)
  • Nutritional factors (alcohol abuse, smoking, eating disorders)
  • Neurologic disease or its treatment
  • Transplantation
  • Genetic metabolic disorders
  • Malignancy.

As in the scenarios presented above, unexplained changes in bone mineral density and mild anemia may trigger an evaluation for a monoclonal gammopathy.


Multiple myeloma is a cancer of the immunoglobulin-producing plasma cells in the bone marrow. Since the cancerous cells are clones, they all produce the same immunoglobulin—thus, the distinctive M-spike on serum or urine protein electrophoresis. It affects about 50,000 people in the United States.

The typical features of multiple myeloma are hypercalcemia, renal insufficiency, anemia, and bone lesions with or without osteoporosis. 11 Most patients have identifiable features of myeloma at the time of diagnosis, but perhaps 20% lack the characteristic symptoms of fatigue, back pain, or bone pain.

Most patients who eventually develop symptomatic multiple myeloma first present with monoclonal gammopathy of undetermined significance (MGUS), a disorder characterized by asymptomatic overproduction of an immunoglobulin. However, MGUS develops into multiple myeloma in only about 15% of cases.11

Widespread osteoporosis, due to cytokine-mediated osteoclast activation, is common in patients with multiple myeloma. As many as 90% of patients have lytic skeletal lesions or osteoporosis at the time of diagnosis.11,12

Myeloma-related osteoporosis can be difficult to differentiate from primary osteoporosis because not all patients secrete a monoclonal protein that standard urine or serum tests can detect.13 But new assays for serum free light chains can help resolve this diagnostic dilemma.14

Next Article: