Aqueous cytomegalovirus DNA and interleukin-8 determination in management of cytomegalovirus retinitis.

Abstract

Cytomegalovirus (CMV) retinitis is the most frequent ocular opportunistic infection after hematopoietic stem cell transplantation (HSCT) (Ochs et al. 1995). For now, the diagnosis of cytomegalovirus retinitis (CMVR) after HSCT is mainly based on clinical history and ophthalmic examination (Port et al. 2017). The traditional treating end-point is also depended on clinical findings, usually described as ‘complete quiescence’ of lesion on the retina in literature (Choopong et al. 2016). But in clinical practice, signs acquired by fundus examination are not always in accordance with actual state of the disease. Here, we conducted an assessment of an antiviral strategy for CMVR after HSCT, using aqueous CMV-DNA load and interleukin-8 (IL-8) level as quantitative variables. Prospective consecutive series of nine patients (15 eyes) who diagnosed as CMVR base on clinical history and fundus examination were given systemic and intravitreous injection of antiviral drugs. Intravitreous injection of ganciclovir (IVG) 3 mg was given to every patient, twice per week for four times (induction phase). Aqueous and serum sample was collected 1 day before the first injection (baseline), during the fourth time of injection, and then every time since the fifth injection. Serum CMV-DNA load, aqueous CMV-DNA load and aqueous IL-8 level were determined by real-time PCR and cytometric beads array, respectively. Compared with baseline, if aqueous CMV-DNA load decreased by two magnitudes or IL-8 level decreased by one magnitude acquired during the fourth injection, IVG 3 mg would continue to be given weekly (maintenance phase), otherwise, therapy would be adjusted to IVG 3 mg combined with foscarnet sodium 2.4 mg once per week (maintenance phase). Local treating end-point was defined as negative aqueous CMV-DNA load (<1.0 × 103 copies/ml) or IL-8 level (<30 pg/ml). Then patients were requested to come back every other week and followed up for 3 months at least. Systemic treating end-point was defined as negative serum CMV-DNA after local treating end-point had been reached. Best corrected visual acuity (BCVA, in form of Logarithm of the minimum angle of resolution, LogMAR), intraocular pressure (IOP; Goldmann applanation tonometry), ophthalmic examination using slit-lamp microscope were performed during each time of visit. Patients were evaluated and followed up for a median of 5 months (3–7 months). Best corrected visual acuity (BCVA) and IOP at base line and last time of visit were (0.679 ± 0.552; LogMAR) and (0.641 ± 0.609; LogMAR), (17.2 ± 5.9) mmHg and (15.9 ± 5.9) mmHg, respectively. No significant difference was found between baseline and last time of visit (p = 0.554 and p = 0.864, respectively). The median times of intravitreous injection were six times (4–11 times). At local treating end-point, systemic antiviral drugs were stopped for all patients, but vessel infiltration, exudation and haemorrhage still could be found in 13 of 15 eyes during fundus examination. Negative aqueous CMV-DNA was found in six eyes (40%), but CMV-DNA could still be detected in nine eyes (60%), while aqueous IL-8 was negative. Median follow-up period was 5 months (3–7 months). At last time of visit, all CMVR lesions became quiescent completely, without any signs of recurrence. Our data indicated that, none of the nine patients experienced recurrence at the last visit, even if CMV-DNA still could be detected in nine eyes at local treating end-point. Furthermore, lesion on the retina continued healing until complete quiescent was achieved even after both systemic and local antiviral therapies were stopped, which proved the efficacy and safety of this strategy. If local treating end-point was set as ‘negative aqueous CMV-DNA’ or ‘complete quiescent lesions on retina’, which is described in literature, the treatment duration must be much longer than actual needs. In summary, aqueous CMV-DNA and IL-8 are excellent quantitative variables that could be used to adjust local antiviral treatment and provide local treating end-point effectively, safely and efficiently.