There’s a new way to evaluate data pooled from many studies when comparing two different treatment techniques for the same problem. The statistical method is called number needed to harm (NNH).
Number Needed to Harm refers to the number of patients treated in a certain way before someone gets hurt or has a complication. It might seem like an odd way to look at something. But using the Number Needed to Harm is more meaningful to a surgeon than using mathematical ratios or odds.
Using the Number Needed to Harm gives the surgeon a better perspective of the benefits and risks of a procedure. Likewise, it gives patients a way to make an informed choice about surgery. This study gives an example of how this statistical method can be used.
Patients with a diagnosis of displaced supracondylar humeral fracture were included in the study. This type of elbow fracture is common in children who reach an arm out to brace themselves when falling. Children involved in this study were between the ages of three and 12.
The upper arm bone (humerus) is fractured above the elbow. The fracture affects the bone just above the natural flare that occurs at the bottom of the bone on both sides. The term displaced tells us the fractured piece of bone has shifted away from the rest of the humerus.
In cases of a displaced supracondylar fracture of the humerus, the displaced piece of bone must be reattached. A wire (called a K-wire) is used to tie the pieces of bone together. The wire can be inserted through the skin and into the bone in a procedure called percutaneous wire fixation. Percutaneous means through the skin.
A percutaneous procedure is considered a closed fixation technique since an incision isn’t needed. Two other ways to perform this operation are open (with an incision) and mini-open (using a small incision).
The surgeon can place one or more wires in one of two ways: straight or what’s referred to as a lateral or crossed pin configuration. As you might imagine, crossing the wire increases the risk of injuring the ulnar nerve, which runs down along the elbow in this area. The question in this study was: which wiring method has the greatest risk of nerve injury?
The authors searched all the published studies they could find for studies that compared crossed and lateral wiring techniques for this type of elbow fracture. They found 32 studies with a total of 2639 patients. Using the Number Needed to Harm, they calculated that the crossed pin method increased the risk of nerve injury. But by how much?
The results showed that for every 28 patients who were treated with the crossed pin fixation method, one would have an ulnar nerve injury. This was the number of patients who would have to be treated by crossed pinning to have one more nerve injury compared with the lateral pin method.
Using the Number Needed to Harm (NNH) as the only measure to show that one method is safer than another is not necessarily recommended. Together with other measuring methods, it provides some additional information for patients and surgeons.
The authors share at the end of their article that there were some challenges in the analysis that might affect the results. First, there were a low number of ulnar nerve injuries with either pinning fixation method. Second, the studies they included in the Systematic Review were retrospective (after the fact). This means there could have been patients who developed ulnar nerve problems later that weren’t included in the study.
Third, the studies included in the review were gathered from 1966 to the present. Many things have changed in surgical technique since the ’60s. It’s possible (even likely) that improved methods have changed the number of nerve injuries reported from study to study. And that could skew the results reported here.
And finally, there were studies that didn’t report the number of ulnar nerve injuries with these two pinning techniques. Patients in the studies had the surgery using the pinning methods but nerve injuries either didn’t occur or weren’t a focus of the study. In either case, that means there were studies on the subject (and a number of patients in those studies) that weren’t included in the total number used to calculate Number Needed to Harm.
The authors conclude that crossed pinning of supracondylar elbow fractures in children does increase the risk of ulnar nerve injury. Surgeons may still opt to use this method because it gives a more stable fixation needed in active children. They must weigh the pros and cons of each technique with each child when making the decision to choose one fixation method over another.