Nonunion is a serious and often undertreated complication that occurs when bone fractures fail to heal properly. About 5-10% of surgically treated fractures experience healing difficulties. Fractures are a common condition, affecting approximately 40% of people at some point in their lives. In Denmark alone, around 15,000 fracture surgeries are performed each year, making fracture treatment one of the largest surgical fields. The impact of nonunion extends beyond the obvious pain and discomfort experienced by patients. There are also significant socioeconomic consequences, such as extended hospital stays, additional surgeries, and prolonged inability to work. Due to the varied nature and relative rarity of nonunion, studying its risk factors requires large amounts of data. Unfortunately, the current literature leaves many important questions unanswered, particularly regarding the actual rate of reoperations due to nonunion after surgical fracture treatment and the risk factors involved. This PhD thesis is comprised of three studies aimed at addressing these gaps.Study I – “Risk factors for nonunion following surgically managed, traumatic, diaphyseal fractures: a systematic review and meta-analysis”The aim of this study was to systematically review the existing literature to identify known risk factors for nonunion after surgically managed, traumatic diaphyseal fractures. A comprehensive literature search was conducted, with studies independently screened by two authors. Data from eligible studies were then extracted. Each study was critically appraised, and random-effects meta-analyses were performed. Out of 11,738 records screened, 30 studies met the eligibility criteria, including a total of 38,465 patients. The analysis found that nonunion was associated with several factors, including smoking, open fractures, diabetes, infection, obesity, higher Gustilo classification, and AO classification. However, the studies reviewed were generally of poor qualityStudy II – “Validation and algorithm development for identifying causes of reoperations after orthopedic fracture surgery: A diagnostic accuracy study using the Danish National Patient Register” Our aim was to validate a method for identifying secondary surgical procedures (reoperations) in the Danish National Patient Register (DNPR) within the first year after initial fracture surgery. The secondary aim was to assess the accuracy of the diagnosis and procedure codes used to determine the causes of these reoperations. Lastly, we created algorithms to improve the precision in identifying the reasons for these reoperations. We used health administrative data from the Danish National Patient Register. The validity assessment was centered on accurately identifying the reasons for reoperations, which we categorized into infections, nonunion, reosteosynthesis, major and minor reoperations, wound complications, and hardware removal. Patient medical records served as our gold standard for comparison. Our findings showed that 94% of the reoperations identified by our algorithm were indeed true reoperations. However, the sensitivity of our classifications varied significantly depending on the type of reoperation, ranging from 23% to 99%. Similarly, the positive predictive value (PPV) also varied widely, from 31% to 100%, depending on the specific algorithm and the reason for the reoperation. Study III – “Distribution and time trends of nonunion following surgically treated extremity fractures: A 20-year nationwide analysis of 340,001 fractures using health administrative data from the Danish National Patient Register” The aim of the study was to explore the demographic differences between patients who developed nonunion and those who did not, focusing on variables such as age, sex, fracture location, primary surgical treatment, and time to treatment. We utilized nationwide health administrative data spanning twenty years. The primary cohort included patients with surgically treated extremity fractures from 1997 to 2016, as recorded in the Danish National Patient Register (DNPR) with both diagnosis and procedure codes. Patients who developed nonunion requiring additional surgery within two years were identified using a previously validated algorithm from study two. In the DNPR, we found 340,001 surgically treated extremity fractures, with 6,928 cases (2.0%) resulting in nonunion. Nonunion was most common among individuals aged 40-49 years, with the lower leg accounting for 31% of all reoperated nonunions. Interestingly, we also discovered that younger individuals with femur fractures had the highest rate of nonunion. Nonunion occurrence decreased significantly during the study period.

Knogleheling kan være en kompleks og udfordrende proces, og i nogle tilfælde heler knoglerne ikke som forventet. Mellem 5-10% af de patienter, der opereres for knoglebrud, oplever helingsproblemer. Formålet med dette projekt er at identificere de risikofaktorer, der bidrager til manglende knogleheling. Knoglebrud er en almindelig tilstand, som cirka 40% af befolkningen vil opleve mindst én gang i løbet af deres liv. Med omkring 15.000 operationer årligt er behandlingen af knoglebrud det største kirurgiske område i Danmark. Problemer med knogleheling kan føre til nedsat funktionsevne, kroniske smerter, økonomiske tab og nedsat arbejdsevne. Derudover er der også socioøkonomiske konsekvenser, såsom længere hospitalsophold, sygemeldinger og gentagne operationer. På trods af problemets omfang er forskningen på området stadig utilstrækkelig. Den nuværende litteratur er sparsom og ofte baseret på små patientgrupper. Denne ph.d.-afhandling består af tre studier:Studie I – “Risk factors for nonunion following surgically managed, traumatic, diaphyseal fractures: a systematic review and meta-analysis”Formålet med dette studie var at systematisk gennemgå den nuværende litteratur for kendte risikofaktorer for nonunion efter kirurgisk behandling af traumatiske diafysære frakturer. En omfattende litteratursøgning blev gennemført, hvor artiklerne blev screenet uafhængigt af to forfattere. Data fra de kvalificerede studier blev derefter udtrukket. Kvaliteten af studierne blev vurderet ved hjælp af et online værktøj, og random effects meta-analyser blev udført. Ud af 11.738 screenede studier var 30 kvalificerede og omfattede i alt 38.465 patienter. Resultaterne viste, at nonunion var forbundet med faktorer som rygning, åbne frakturer, diabetes, infektion, fedme samt stigende Gustilo- og AO-klassifikation. Studierne, der blev gennemgået, var generelt af lav kvalitet.Studie II – “Validation and algorithm development for identifying causes of reoperations after orthopedic fracture surgery: A diagnostic accuracy study using the Danish National Patient Register”Formålet med studiet var at validere en metode til at identificere sekundære kirurgiske indgreb (reoperationer) i Landspatientregisteret inden for det første år efter den primære frakturoperation. Det sekundære mål var at vurdere nøjagtigheden af de diagnose- og procedurekoder, der blev brugt til at fastslå årsagerne til disse reoperationer. Endelig udviklede vi algoritmer for at øge præcisionen i at identificere årsagerne til reoperationerne. Validitetsvurderingen fokuserede på at bestemme de reelle årsager til reoperationer, som blev kategoriseret i infektioner, nonunion, reosteosyntese, større og mindre reoperationer, sårkomplikationer og fjernelse af implantater. Patientjournaler blev brugt som guldstandard til sammenligning. Resultaterne viste, at 94% af de identificerede reoperationer faktisk var reoperationer. Sensitiviteten af vores klassifikationer varierede dog afhængigt af typen af reoperation, fra 23% til 99%. Den positive prædiktive værdi varierede også betydeligt, fra 31% til 100%, afhængigt af den specifikke algoritme og årsagen til reoperationen.Studie III – “Nonunion in surgically treated patients: A 20-year nationwide analysis of 340,001 extremity fractures using health administrative data from the Danish National Patient Register”Studiet havde til formål at undersøge de demografiske forskelle mellem patienter med nonunion og dem uden, med fokus på faktorer som alder, køn, frakturens placering, primær kirurgisk behandling, og tiden fra skade til behandling. Vi benyttede os af landsdækkende sundhedsdata over en periode på tyve år. Den primære kohorte bestod af patienter med ekstremitetsfrakturer, der blev behandlet kirurgisk med osteosyntese fra 1997 til 2016 og registreret i Landspatientregisteret (DNPR) med diagnose- og procedurekoder. Nonunions blev identificeret ved hjælp af en tidligere valideret algoritme fra studie to. Vi fandt 340.001 kirurgisk behandlede ekstremitetsfrakturer i DNPR, hvoraf 6.928 (2,0%) udviklede nonunion. Nonunion var mest udbredt blandt personer i alderen 40-49 år, og underbenet udgjorde 31% af alle reopererede nonunions. Overraskende nok viste resultaterne, at unge mennesker med lårbensbrud havde den højeste forekomst af nonunion. Derudover fandt vi, at antallet af kirurgisk behandlede nonunions faldt signifikant i løbet af undersøgelsesperioden.