Explore the vast range of titles available.

This study, designed to determine the relative degree of testosterone deficiency, estradiol deficiency, or both at which undesirable bodily changes occur, showed that some features of male hypogonadism are due to both androgen deficiency and estrogen deficiency. Testosterone therapy is prescribed for millions of men each year, and the number is increasing rapidly. Prescription sales of testosterone increased by 500% in the United States between 1993 and 2000. 1 Most testosterone prescriptions are written to treat nonspecific symptoms, such as fatigue or sexual dysfunction, when accompanied by testosterone levels below the laboratory reference range. Currently, testosterone levels that are at least 2 SD below the mean value for healthy young adults are classified as low. 1 , 2 Although convenient, this classification fails to consider the physiological consequences of specific testosterone levels. More than 80% of circulating estradiol in men . . .

This interactive features a 63-year-old woman with a history of multiple fractures who presents with progressively worsening pain in both ankles and difficulty walking. Test your diagnostic and therapeutic skills at NEJM.org.

A 45-year-old woman presented with hypertension, fatigue, and episodic confusion. After medications were administered, the blood pressure decreased but fatigue and confusion persisted. Four weeks later, pulmonary embolism developed. A diagnostic test was performed.

A complex homeostatic system involving the interplay of bone, the kidneys, and the intestines maintains extracellular calcium levels within a relatively narrow range. This article describes a patient with autoimmune hyperparathyroidism and hypocalciuric hypercalcemia caused by IgG4 autoantibodies directed against the calcium-sensing receptor. The patient's hypercalcemia and elevated parathyroid hormone levels responded to the administration of glucocorticoids. IgG4 autoantibodies and a novel mechanism resulting in inactivating mutations of the calcium-sensing receptor. A complex homeostatic system involving the interplay of the bones, the kidneys, and the intestines has evolved to maintain extracellular calcium concentrations within a relatively narrow range. 1 The primary regulator of this system is parathyroid hormone, the release of which is initiated by signals from the calcium-sensing receptor. Overproduction of parathyroid hormone gives rise to hypercalcemia by stimulating the efflux of calcium from bone, increasing the reabsorption of urinary calcium, and promoting the uptake of dietary calcium by means of the activation of vitamin D. 1 Parathyroid hormone–dependent hypercalcemia is commonly caused by parathyroid adenomas and hyperplasia. 2 Rarer causes of parathyroid . . .

To the Editor: In the Case Record, Pallais et al. (Oct. 25 issue) 1 recommend the measurement of proinsulin to help distinguish an insulinoma from secretagogue-induced insulin release. However, proinsulin is important for identifying insulinomas that predominantly produce and secrete proinsulin. We examined a 61-year-old man with episodic sweating, tremor, and confusion that resolved with carbohydrate ingestion. The postabsorptive serum glucose level was 50 mg per deciliter (2.8 mmol per liter) and the insulin level was 10 μU per milliliter (69.4 pmol per liter). On another occasion, the postabsorptive glucose level was 81 mg per deciliter (4.5 mmol per liter), the . . .

Abstract Disclosure: B.E. Bertot: None. J. Samsel: None. J. Pallais: Alexion Pharmaceuticals, Inc., Research, UpToDate. Hypophosphatasia (HPP) is an underrecognized genetic disorder caused by loss-of-function mutations in ALPL, the gene encoding tissue-nonspecific alkaline phosphatase (TNAP), including the liver and bone isoforms. Current biochemical diagnostic strategies rely on persistently low total alkaline phosphatase (ALP) levels along with elevated ALP substrates such as pyridoxal 5’-phosphate (vitamin B6) and phosphoethanolamine (PEA). However, these tests demonstrate limited sensitivity, and many patients with genetically confirmed HPP are missed. We investigated whether measurement of the liver 1 isoform of ALP, using gel electrophoresis, enhances diagnostic sensitivity. Methods: We analyzed 22 adults with genetically or clinically confirmed HPP who underwent fractionated ALP testing by electrophoresis. To minimize diagnostic ambiguity, we calculated strict sensitivity—defined as all values for a given biomarker falling within the abnormal range expected in HPP—for total ALP, vitamin B6, urine PEA, bone-specific ALP (by immunoassay), and fractionated liver and bone ALP isoforms. We then assessed how the addition of liver 1 ALP isoform data affected sensitivity and tested differences using paired proportion comparisons. Results: The liver 1 ALP isoform had the highest strict sensitivity of any individual test (68.2%). Other tests showed lower sensitivities: total ALP (40.9%), vitamin B6 (36.4%), PEA (47.4%), bone-specific ALP (38.1%), and fractionated bone ALP (31.8%). Adding liver 1 ALP significantly improved sensitivity for each of these tests:•Vitamin B6 + liver 1 ALP: 90.9% (Δ +54.5%, p=0.001)•Fractionated bone + liver 1 ALP: 72.7% (Δ +40.9%, p=0.007)•Bone-specific ALP + liver 1 ALP: 76.2% (Δ +38.1%, p=0.013)•PEA + liver 1 ALP: 78.9% (Δ +31.6%, p=0.044)•Total ALP + liver 1 ALP: 68.2% (Δ +27.3%, p=0.069). Among patients missed by each individual test, liver 1 accurately identified 85.7% of those missed by B6, 60% by PEA and fractionated bone, and 46.2% by total ALP. Heat map analysis confirmed liver 1 ALP isoform as the most effective additive marker, consistently enhancing the performance of all standard tests. Conclusion: Measurement of the liver 1 ALP isoform by electrophoresis significantly improves the diagnostic sensitivity of standard biochemical tests for HPP. Its inclusion in clinical testing protocols may reduce underdiagnosis in adults suspected of having HPP. Presentation: Sunday, July 13, 2025

Case 7-2011: A 52-Year-Old Man with Upper Respiratory Symptoms and Low Oxygen Saturation Levels Case Records of the Massachusetts General Hospital, N Engl J Med 2011;364:957-966. In the Causes of Low Oxygen Saturation on Pulse Oximetry subsection of Differential Diagnosis, the unit of measure for wavelengths should have been nm, rather than mm, in two instances: in the first sentence under Hypoxemia (page 959) and in the second sentence of the second paragraph under Dapsone and Methemoglobinemia (page 960). The article is correct at NEJM.org.

Case 7-2011: A 52-Year-Old Man with Upper Respiratory Symptoms and Low Oxygen Saturation Levels Case Records of the Massachusetts General Hospital, N Engl J Med 2011;364:957-966. In the legend for Figure 2 (page 964), the penultimate sentence should have read, “The duodenal-biopsy specimen (Panel C, hematoxylin and eosin) shows blunted and atrophic villi (long arrow), crypt hyperplasia (short arrows) . . . ,” rather than “. . . shows blunted and atrophic villi (short arrows), crypt hyperplasia (long arrow). . . .” The article is correct at NEJM.org.

Large language models (LLMs) are routinely used by physicians and patients for medical advice, yet their clinical safety profiles remain poorly characterized. We present NOHARM (Numerous Options Harm Assessment for Risk in Medicine), a benchmark using 100 real primary care-to-specialist consultation cases to measure frequency and severity of harm from LLM-generated medical recommendations. NOHARM covers 10 specialties, with 12,747 expert annotations for 4,249 clinical management options. Across 31 LLMs, potential for severe harm from LLM recommendations occurs in up to 22.2% (95% CI 21.6-22.8%) of cases, with harm of omission accounting for 76.6% (95% CI 76.4-76.8%) of errors. Safety performance is only moderately correlated (r = 0.61-0.64) with existing AI and medical knowledge benchmarks. The best models outperform generalist physicians on safety (mean difference 9.7%, 95% CI 7.0-12.5%), and a diverse multi-agent approach improves safety compared to solo models (mean difference 8.0%, 95% CI 4.0-12.1%). Therefore, despite strong performance on existing evaluations, widely used AI models can produce severely harmful medical advice at nontrivial rates, underscoring clinical safety as a distinct performance dimension necessitating explicit measurement.