Incidence, Prevention, and Management of Complications Following Percutaneous Nephrolitholapaxy

ContextIncidence, prevention, and management of complications of percutaneous nephrolitholapaxy (PNL) still lack consensus.ObjectiveTo review the epidemiology of complications and their prevention and management.Evidence acquisitionA literature review was performed using the PubMed database between 2001 and May 1, 2011, restricted to human species, adults, and the English language. The Medline search used a strategy including medical subject headings (MeSH) and free-text protocols with the keywords percutaneous, nephrolithotomy, PCNL, PNL, urolithiasis, complications, and Clavien, and the MeSH terms nephrostomy, percutaneous/adverse effects, and intraoperative complications or postoperative complications.Evidence synthesisAssessing the epidemiology of complications is difficult because definitions of complications and their management still lack consensus. For a reproducible quality assessment, data should be obtained in a standardized manner, allowing for comparison. An approach is the validated Dindo-modified Clavien system, which was originally reported by seven studies. No deviation from the normal postoperative course (Clavien 0) was observed in 76.7% of PNL procedures. Including deviations from the normal postoperative course without the need for pharmacologic treatment or interventions (Clavien 1) would add up to 88.1%. Clavien 2 complications including blood transfusion and parenteral nutrition occurred in 7%; Clavien 3 complications requiring intervention in 4.1.%; Clavien 4, life-threatening complications, in 0.6%; and Clavien 5, mortality, in 0.04%. High-quality data on complication management of rare but potentially debilitating complications are scarce and consist mainly of case reports.ConclusionsComplications after PNL can be kept to a minimum in experienced hands with the development of new techniques and improved technology. A modified procedure-specific Clavien classification should be established that would need to be validated in prospective trials.