Teen dies when blood culture protocol botched

How a Shoulder Complaint Became a Fatal Infection

The reported case began around 4 a.m., when a 17-year-old awoke with intense pain in his left shoulder. The pain became severe enough for his parents to call 911, and by the time he reached the emergency department, it extended toward his wrist.

There was an easy explanation available. Three days earlier, the teenager had lifted roughly 200 pounds at the gym and noticed soreness afterward. Weightlifting plus shoulder pain seemed to equal a muscle or joint injury. Unfortunately, the rest of the clinical picture did not fit neatly inside that equation.

The teenager also reported a headache, a mild fever, coughing with green sputum, and vomiting. His shoulder movement was limited, and his white blood cell count was approximately 19,000, suggesting that his immune system was reacting to something significant. Chest and shoulder X-rays were normal. Because of the headache, the emergency physician performed a spinal tap, which also returned normal results. Blood cultures were collected to look for bacteria circulating in the bloodstream.

After approximately 11 hours, intravenous pain medicine had reduced his discomfort. His arm was placed in a sling, and he was discharged with a shoulder-sprain diagnosis. He was told to use ibuprofen and visit his primary care physician the next day for follow-up, including the pending blood culture results.

The next morning, the laboratory reportedly notified the emergency department that both cultures were growing the same form of streptococcus. According to the case review, a health unit coordinator passed the report to a charge nurse, who presented it to the emergency physician on duty. The communication was not documented in the medical record, and no treatment or patient recall followed.

That afternoon, the teenager visited his primary care physician. He still had a fever. Suspecting a deep tissue infection, the physician ordered another set of blood cultures and scheduled a shoulder MRI for the following morning. The physician was reportedly unaware that the earlier cultures were already positive. The teenager went home again.

His condition worsened overnight. When he returned to the hospital, an MRI revealed septic arthritis in the shoulder. He was admitted and treated, but the infection ultimately proved fatal.

What “Botched Blood Culture Protocol” Actually Means

The title can create a misleading picture. The central allegation was not necessarily that the blood was drawn incorrectly or that the laboratory failed to grow the organism. In fact, the cultures did exactly what they were supposed to do: they produced an early warning that bacteria might be circulating in the patient’s blood.

The alleged failure occurred after the preliminary result became available. The hospital reportedly had a policy requiring a designated culture nurse to notify an emergency physician, contact the patient’s primary care provider, fax the result, and inform the patient. During litigation, staff members offered conflicting recollections about who received the report, what was said, and whether anyone decided to wait for a final result.

That distinction matters. Collecting a diagnostic test is only the opening move. A test has little safety value unless someone receives the result, understands its significance, documents the communication, makes a clinical decision, and ensures that the decision reaches the patient.

A protocol that lives only in a policy binder is not a functioning safety system. It is office furniture with ambitious vocabulary.

Why Two Positive Cultures Deserved Immediate Attention

Blood cultures help identify bacteria or fungi in the bloodstream. They are especially useful when clinicians suspect bacteremia, a serious localized infection, or sepsis. Because microorganisms need time to multiply inside culture bottles, results often remain pending when a patient leaves the emergency department.

Not every positive result represents a dangerous bloodstream infection. Skin organisms can accidentally enter a culture bottle during collection, producing contamination and potentially leading to unnecessary antibiotics. Clinicians therefore consider the organism, number of positive bottles or sets, time to positivity, symptoms, vital signs, laboratory findings, and likelihood of infection.

However, multiple cultures growing the same organism generally deserve more concern than one isolated bottle containing a common skin contaminant. In this case, both preliminary cultures reportedly grew the same type of streptococcus in a febrile teenager with severe localized pain and a markedly elevated white blood cell count. “Preliminary” did not mean “imaginary.” It meant that complete identification and antibiotic-susceptibility testing were still underway.

AHRQ patient-safety guidance emphasizes that discharged patients with pending microbiology tests need a reliable system that identifies actionable results and promptly notifies the appropriate clinicians. When interpretation is uncertain, infectious-disease consultation may be appropriate rather than simply waiting and hoping the problem becomes clearer on its own.

The Diagnostic Trap: Anchoring on the Gym Injury

The weightlifting history may have created an anchoring effect. Anchoring bias occurs when an early piece of information becomes the dominant explanation and later evidence is interpreted around it. Once “shoulder strain” entered the story, the headache, fever, vomiting, productive cough, leukocytosis, and extreme pain could be treated as unrelated details instead of pieces of one dangerous puzzle.

A lifting injury was certainly possible. Teenagers lift heavy weights, become sore, and occasionally discover muscles they did not know existed. But a routine strain did not fully explain the systemic symptoms or the elevated white blood cell count.

Premature closure may also have played a role. This happens when a clinician accepts an initial diagnosis before sufficiently considering alternatives. A useful safeguard is a “diagnostic time-out” before discharge:

• What findings do not fit the working diagnosis?
• What dangerous condition could produce this presentation?
• Why were blood cultures ordered?
• Who will review them if they become positive?
• Would a negative X-ray exclude the diagnoses that matter most?

AHRQ and the American Medical Association describe anchoring as continued reliance on an initial impression even when later information points elsewhere. Reassessment, explicit consideration of competing diagnoses, and documentation of uncertainty can reduce the risk of that cognitive trap.

Septic Arthritis Can Resemble an Ordinary Injury

Septic arthritis is an infection inside a joint. Bacteria can enter through an injury or procedure, but they often reach the joint through the bloodstream. The knee and hip are more commonly affected, although the shoulder can also become infected.

Symptoms may include rapidly developing joint pain, restricted movement, swelling, warmth, and fever. Yet the presentation is not always textbook-perfect. A patient may have little visible redness, a mild temperature, or normal early X-rays. Plain radiographs are useful for detecting fractures and certain structural problems, but they cannot reliably rule out an early joint infection.

Blood tests such as C-reactive protein and erythrocyte sedimentation rate can support the evaluation, while MRI may reveal infection in the joint, bone, or surrounding tissue. The definitive evaluation commonly includes sampling synovial fluid from the joint for cell counts, Gram staining, and culture.

Septic arthritis is treated urgently because infection can damage cartilage, spread into nearby bone, enter the bloodstream, and trigger sepsis. Medical guidance warns that delayed diagnosis can lead to permanent joint damage, serious systemic illness, and death.

Sepsis Does Not Wait for the Paperwork

Sepsis is the body’s extreme and life-threatening response to infection. It can cause tissue injury, organ dysfunction, shock, and death. Warning signs may include fever or an abnormally low temperature, rapid heart rate, rapid breathing, confusion, clammy skin, unusually severe pain, low blood pressure, and reduced urine output.

There is no single laboratory result that confirms every case. Clinicians combine the patient’s appearance, vital signs, suspected source of infection, organ function, imaging, cultures, and response to treatment. That uncertainty is exactly why repeated assessment matters.

CDC guidance calls for immediate evaluation and treatment when sepsis is suspected. Blood cultures may help identify the organism, but treatment should not be placed on pause while everyone waits for perfect microbiological certainty. Current sepsis guidance recommends rapid antimicrobial treatment for patients with septic shock or a high likelihood of sepsis, while also encouraging prompt assessment to avoid unnecessary antibiotic use in lower-risk cases.

The goal is not to give antibiotics to every person with shoulder pain. The goal is to recognize when pain is accompanied by enough evidence of infection that the risk calculation has changed dramatically.

The Missing Safety Feature: Closed-Loop Communication

In a closed-loop system, the laboratory sends the critical result, a responsible clinician receives it, receipt is acknowledged, the finding is interpreted, an action is selected, the action is documented, and the patient is contacted. The loop is not closed merely because someone says, “I told the nurse,” or because an alert landed in an electronic inbox containing 147 other alerts.

The Joint Commission identifies closed-loop communication as an essential defense against diagnostic error. Hospitals are expected to define which results are critical, who must receive them, how quickly they must be reported, and how the process will be documented and evaluated.

In practical terms, a safer positive-culture workflow should include:

• A clearly named clinician or team responsible for every pending culture.
• Automatic escalation when the first recipient does not acknowledge the result.
• Documentation of the time, recipient, interpretation, and planned response.
• Direct outreach to the patient or guardian using multiple contact methods when necessary.
• Notification of the primary care physician and other relevant clinicians.
• A daily queue showing unresolved results rather than merely “delivered” messages.
• Auditing to confirm that communication led to an appropriate clinical action.

The electronic health record should behave less like a filing cabinet and more like a smoke alarm. Delivery is not the endpoint; verified action is.

Discharging a Patient With Pending Results

Discharging someone before all test results are complete is not automatically unsafe. Cultures, pathology studies, and specialized tests can take days. Keeping every stable patient in the hospital until every result is final would create new risks and overwhelm available beds.

The danger appears when responsibility becomes vague. The emergency physician may assume the primary care doctor will review the result. The primary care doctor may not know the test exists. The laboratory may assume that reporting the value to any staff member completes its duty. The patient may believe that silence means everything was normal.

Before discharge, the medical record and patient instructions should identify pending tests, explain why they matter, state who is responsible for review, provide an expected timeline, and describe how results will be communicated. Patients should also receive specific return precautions rather than the wonderfully unhelpful phrase “come back if worse.”

AHRQ discharge resources recommend deliberate follow-up of pending laboratory results because many require a change in diagnosis or treatment. No patient should need detective skills, three passwords, and a lucky phone call to discover a potentially life-threatening result.

What Families Can Ask Before Leaving the Emergency Department

Patients and families cannotand should notbe expected to repair hospital workflow. Still, a few direct questions can make pending tests and safety plans more visible:

• Are any laboratory, culture, imaging, or specialist results still pending?
• Who is personally responsible for reviewing those results?
• How and when will we be contacted?
• What should we do if no one contacts us?
• Which symptoms mean we should return immediately?
• Could the severe pain and fever indicate a deep infection or sepsis?
• Can we receive written discharge instructions and access the patient portal?

Anyone who develops worsening severe pain, breathing difficulty, confusion, fainting, bluish or mottled skin, very rapid breathing, extreme weakness, or other signs of serious deterioration should seek emergency care immediately rather than waiting for a scheduled appointment or culture update.

Experience-Based Lessons From Similar Patient-Safety Events

The Family’s Experience: “We Followed the Instructions”

One of the most painful features of cases involving pending results is that families often do exactly what they were told. In this reported case, the parents called 911 when the pain became severe, remained through a lengthy emergency evaluation, took their son home after discharge, and brought him to his primary care physician the next day. They did not ignore the medical plan.

After a bad outcome, families may repeatedly revisit every decision: Should we have refused discharge? Should we have demanded another scan? Should we have returned sooner? That hindsight can create enormous guilt, even though the family did not have access to the positive culture result or the clinical knowledge needed to interpret it. A well-designed system must not depend on parents correctly guessing that the diagnosis is wrong.

The Clinician’s Experience: A Familiar Story Becomes Too Convincing

Emergency clinicians work in an environment filled with interruptions, uncertainty, crowding, and incomplete information. A young person reporting pain after heavy lifting presents a familiar and statistically reasonable story. The danger arises when familiarity reduces curiosity.

Experienced clinicians often describe the value of noticing “friction” in a casethe detail that keeps rubbing against the diagnosis. Here, severe pain requiring an ambulance, systemic symptoms, restricted shoulder motion, leukocytosis, and the decision to order blood cultures all created friction against a simple sprain. A diagnostic time-out does not require an hour-long conference. Sometimes it is a 30-second question: “What serious diagnosis am I about to send home?”

The Nurse and Laboratory Experience: Reporting Is Not the Same as Resolving

Laboratory staff may correctly phone a preliminary result, and a nurse may correctly repeat it to a physician, yet the patient can remain at risk if nobody records or owns the next step. Similar safety investigations frequently reveal a chain of individually plausible actions that collectively produces failure.

This is why read-back and acknowledgment matter. The receiver should repeat the result, document it, and state the plan. If the plan is to await final identification, the reason should be clinically justified and visible. If the patient has been discharged, a named person should confirm successful contact. “Message sent” is an administrative status. “Patient evaluated and treatment arranged” is a clinical outcome.

The Hospital Leader’s Experience: Policy Without Measurement Is Wishful Thinking

Hospitals commonly respond to adverse events by revising a policy or providing more education. Those steps may help, but they are weak if leaders do not measure whether the process works during busy nights, weekends, shift changes, staffing shortages, and electronic downtime.

A stronger approach audits a sample of positive cultures from discharged patients. Leaders can examine how long notification took, whether the clinician acknowledged it, whether the patient was reached, and how quickly treatment changed. They can also simulate a critical result after hours to see where the handoff fails. Safety procedures should be tested under real-world pressure, not admired under conference-room lighting.

The Patient-Safety Experience: Build for Human Fallibility

The most durable lesson is that errors rarely belong to one person alone. A clinician may anchor on an injury, a nurse may assume the doctor will act, a coordinator may believe the handoff is complete, and an electronic system may display the result without demanding resolution. Each gap aligns with the next.

Health care cannot eliminate fatigue, distraction, uncertainty, or imperfect memory. It can create systems that expect those human limitations. Redundant alerts, clear ownership, escalation rules, shared documentation, patient notification, diagnostic time-outs, and retrospective audits provide multiple opportunities to catch a mistake before it becomes irreversible.

Conclusion: A Result Matters Only When It Changes Care

The death of a teenager after an initially missed joint infection is not simply a story about one diagnosis. It is a warning about how diagnostic bias and broken communication can amplify each other.

The shoulder pain was attributed to weightlifting. The systemic clues did not receive enough weight. Blood cultures were ordered, but the responsibility for following them became fragmented. Two preliminary positive results reportedly reached the emergency department without producing documented action. The primary care physician evaluated the teenager without access to the information that might have changed the plan.

Reliable care requires more than talented individuals. It requires a system in which dangerous findings cannot quietly disappear between the laboratory, nursing station, physician inbox, primary care office, and patient’s home.

Blood cultures can identify a hidden infection, but they cannot call a family, arrange an admission, begin antibiotics, or document a decision. People and systems must complete those steps. When the warning is serious, the loop must close quicklybecause bacteria do not pause for shift change.

Note: This educational article is based on an anonymized physician case analysis and guidance from U.S. patient-safety and medical organizations, including KevinMD, the CDC, AHRQ Patient Safety Network, the Joint Commission, the Society of Critical Care Medicine, the American Academy of Family Physicians, the American Society for Microbiology, the Infectious Diseases Society of America, NIH/NCBI, MedlinePlus, and the American Medical Association. It does not provide a legal conclusion or replace individual medical evaluation.