The diagnostic performance of dual phase [99mTc]-MIBI SPECT/CT in primary hyperparathyroidism assessment: limits and prediction model of false negative result considering pre-scan data in a single center experience

The human body relies on a delicate balance of minerals to keep bones strong and nerves firing correctly. One of the key regulators of this balance is a small gland located in the neck, just behind the thyroid, known as the parathyroid. When this gland becomes overactive, it pumps out too much of a hormone called parathyroid hormone. This condition, known as primary hyperparathyroidism, causes the body to release too much calcium from the bones into the bloodstream. Over time, this excess calcium can lead to kidney stones, bone thinning, and fatigue. While the condition is common and often manageable, the most effective cure is usually a surgical removal of the overactive gland. To perform this surgery safely and precisely, doctors must first locate the rogue gland, which can be as small as a pea and sometimes hides in unusual places within the neck or chest.

Locating this tiny gland is a significant challenge. Doctors typically use two main tools: an ultrasound, which uses sound waves to create an image, and a specialized nuclear medicine scan called a Sestamibi scan. In this scan, a patient receives a tiny amount of a radioactive tracer that is attracted to the overactive gland, lighting it up on a camera. However, this method is not perfect. In roughly one out of every four cases, the scan fails to find the gland, leaving the surgeon without a clear map. When this happens, patients often need to undergo more expensive and complex imaging tests, or they may face a more difficult surgery where the surgeon has to search blindly. The question of why the scan sometimes fails, and whether doctors could predict this failure before the scan is even done, has remained a difficult puzzle.

A team of researchers at the ASST Spedali Civili in Brescia, Italy, set out to solve this puzzle by looking back at the medical records of 64 patients who had undergone this specific scanning procedure. They wanted to see if information already available before the scan—such as blood test results and the findings from a neck ultrasound—could predict whether the scan would successfully find the gland or miss it. The team gathered data on the patients' calcium levels, their parathyroid hormone levels, and the size and location of any glands spotted by ultrasound. They then used a statistical method to see if these pre-scan details were linked to the final outcome of the imaging.

The researchers discovered that the results of the scan were closely tied to two specific factors: the level of calcium in the blood and the findings from the neck ultrasound. Patients who had higher calcium levels and a positive ultrasound result were much more likely to have a successful scan that correctly identified the overactive gland. Conversely, patients with lower calcium levels or a negative ultrasound were more likely to experience a false negative, where the scan missed the gland entirely. The study also noted that the size of the gland mattered; larger glands were easier to spot. Interestingly, while the hormone levels were high in patients with successful scans, the researchers found that calcium levels and ultrasound results were the strongest predictors when used together.

To make these findings useful for daily practice, the team built a mathematical model designed to estimate the risk of a missed diagnosis for any given patient. By plugging in a patient's calcium level and ultrasound result, the model could calculate a probability score. In their testing, this approach correctly identified the outcome in about 72 percent of the cases. This means that for a significant portion of patients, doctors could potentially know in advance if the standard scan is likely to fail. If the model suggests a high risk of a missed gland, a doctor might decide to skip the standard scan and go straight to a more sensitive, albeit more expensive, imaging test. This could save time, reduce the number of procedures a patient undergoes, and help surgeons plan their approach with greater confidence.

The study concludes that while this predictive tool is promising, it is not yet a final solution. The group of patients they studied was relatively small, and the model needs to be tested on a larger, more diverse group of people to ensure it works for everyone. The authors emphasize that their work is a step toward a more personalized approach to diagnosis, where decisions are guided by a patient's specific data rather than a one-size-fits-all protocol. By understanding the limits of current imaging and using simple blood and ultrasound data to anticipate those limits, doctors can better navigate the path to a cure for this common condition. The ultimate goal is to ensure that every patient receives the right test at the right time, avoiding unnecessary delays and procedures while securing the best possible outcome.