Atlas of Procedures in Neonatology, 4th Edition

Miscellaneous Sampling


Ophthalmic Specimen Collection

Jennifer A. Dunbar

  1. Indications
  2. To obtain specimen for testing to determine the cause of conjunctivitis (Table 23.1)
  3. Neonatal conjunctivitis is considered an ocular emergency (1).
  4. Neonatal conjunctivitis is defined as any conjunctivitis presenting in an infant during the first 28 days of life (2).
  5. Signs and symptoms include diffuse conjunctival injection with mucoid, purulent, or watery ophthalmic discharge.
  6. Conjunctivitis may be the presenting sign of coexisting life-threatening systemic infection.
  7. Both bacterial and viral pathogens cause corneal ulceration and opacity, which may lead to blindness. Neisseria gonorrheaorPseudomonasspecies may rapidly perforate the globe.
  8. To test for colonization when a conjunctivitis epidemic is suspected in the neonatal intensive care unit
  9. Conjunctivitis is common in the neonatal intensive care unit, affecting up to 6% of infants (3).
  10. The eye may be contaminated by respiratory secretions, with coagulase-negative Staphylococcus, S. aureus, and Klebsiellaspecies reported as the most common pathogens. These are associated with ventilator or nasal continuous positive airway pressure (3).
  11. Conjunctivitis epidemics have been associated with routine ophthalmic screening in the neonatal intensive care unit (4).
  12. Serratia marcessens, Klebsiellaspecies, and Adenovirusepidemics have been described (4,5).
  13. Systemic illness and death have been associated with conjunctivitis epidemics in the neonatal intensive care unit (4,5).

Because conjunctivitis of the newborn is often a sign of a maternally transmitted systemic infection, work-up includes maternal history of vaginal symptoms or discharge and maternal vaginal culture results (6).

  1. Relative Contraindications
  2. Corneal epithelial defect
  3. If fluorescein staining of the cornea reveals an epithelial staining defect, then corneal ulceration may be present. This requires referral to an ophthalmologist.
  4. Special Considerations for Ophthalmic Specimen Management
  5. Conjunctival scrapings are the specimen of choice because many pathogens are intraepithelial (7).
  6. The ocular specimen size is small; therefore, special care is given to specimen handling.
  7. Direct placement of the conjunctival scrapings on slides for staining and direct plating onto culture medium at the bedside will maximize the yield.
  8. Communication with laboratory personnel regarding specimen handling improves culture results (8).
  9. Materials
  10. Equipment for staining the cornea to rule out epithelial defect
  11. Fluorescein dye or strips
  12. Wood lamp or other blue light source
  13. Equipment for obtaining specimen
  14. Choose topical anesthetic (optional):
  15. 0.5% preservative-free tetracaine in unit-dose containers (Alcon Laboratories, Fort Worth, TX, USA)
  16. Preservative-free lidocaine (Elkins-Sinn, Cherry Hill, NJ, USA)
  17. Cocaine 4% preserved with 0.5% sodium benzoate (Schein Pharmaceuticals, Port Washington, NY, USA) diluted to 1% to 2% with sterile water.

Many topical ophthalmic anesthetics contain preservatives that may inhibit bacterial growth in culture. For this reason, some physicians choose to perform the procedure without anesthetic. Nevertheless, this may be quite painful for the infant. The above-mentioned


anesthetics are preservative-free or have been shown not to inhibit bacterial growth (9).

  1. Sterile cotton swabs may be used to evert the eyelids but are not recommended for specimen collection.
  2. Choose instrument to obtain cultures:
  3. Calcium alginate swabs
  4. Sterile Dacron polyester-tipped applicator (Harwood Products Company, Guilford, ME, USA)

Calcium alginate swabs have been shown to yield equal or better organism retrieval in cultures than spatulas or Dacron swabs (10,11). Moistening the swab with tripticase soy (Becton Dickenson and Company, Cockeysville, MD, USA) broth or other culture medium enhances results. However, spatulas have been shown to provide better samples in smear than swabs. Spatulas preserve the conjunctival epithelial cells better, thus providing better opportunity for diagnosing pathogens with intracellular organisms or inclusions (12). Calcium alginate swabs may interfere with immunoassays.

  1. Choose instrument for scraping the conjunctiva:
  2. Kimura Platinum E-109 spatula (Storz Instrument Co., St. Louis, MO, USA)
  3. Nasopharyngeal swab with metal handle bent for scraping
  4. Calcium alginate swabs

If spatulas are not available, then swabs should be used vigorously on the tarsal conjunctival surface so as to débride epithelial cells.

  1. Equipment for obtaining microscope slides
  2. Frosted, etched glass slides
  3. Microslide holders
  4. Pencil or marker for labeling

TABLE 23.1 Analysis of Conjunctival Scrapings


Organisms Identified



Gram's stain

Neisseria gonorrhea

Gram-negative diplococci

Giemsa stain

Chlamydia trachomatis

Intraepithelial intracytoplasmic inclusions

Papanicolaou stain

Herpes simplex virus

Multinucleate giant cells and inclusion-bearing cells

Direct antigen detection techniques

Immunofluorescent indicator system

Chlamydia trachomatis

Immunosorbent assay (ELISA)

Chlamydia trachomatis
Herpes simplex virus

Fluorescein-labeled monoclonal antibodies (Syva MicroTrack)

Chlamydia trachomatis

Indirect fluorescence

Herpes simplex virus


Specific media


Neisseria gonorrhea


Gram-positive and gramnegative bacteria


Anaerobic bacteria

Viral transport

Herpes simplex virus

Chlamydia culture (McCoy culture)

Chlamydia trachomatis

ELISA, enzyme-linked immunosorbent assay.

  1. Equipment for Identifying Chlamydia
  2. Equipment for nonculture chlamydial studies

Traditionally, McCoy culture was considered the “gold standard” for identification of Chlamydia. However, cultures take several days to provide results, and specimens collected in the first few days of life may have less yield on culture because elementary bodies often take several days to form in neonates (13,14). The nonculture tests listed below, such as direct immunofluorescence, enzyme-linked immunosorbent assay, and polymerase chain reaction (PCR), all perform well for ocular specimens and provide a result more rapidly. Polymerase chain reaction is now commercially available and is probably the most sensitive nonculture test for Chlamydia available (14). Real-time PCR is becoming available for ophthalmic specimens when Chlamydia and/or viral


infections such as herpes virus are suspected (15,16). Check with the laboratory involved regarding validation of PCR for ophthalmic specimens.

  1. For chlamydial DFA stain: Syva MicroTrak Chlamydia trachomatisspecimen collection kit (Trinity Biotech, Co., Wicklow, Ireland)
  2. For chlamydial enzyme-linked immunosorbent assay: Place specimen in media advised by the laboratory performing the study.
  3. For chlamydial polymerase chain reaction: Place specimen in transport medium appropriate for the assay used (13). An example is M4 medium for the transport of viruses and Chlamydia(Remel, Lenexa, KS, USA).
  4. Media for Chlamydiaculture

Specimens should be plated onto culture medium at the bedside. Each laboratory will have specific media available for a particular type of organism. Check with the laboratory involved for the appropriate medium for that hospital. The following list is a suggestion of classic media used for each type of organism.

  1. Bacterial culture media
  2. Trypticase soy broth
  3. Blood agar plate
  4. Chocolate agar plate for Hemophilus influenzae, Neisseria gonorrhea
  5. Thayer-Martin medium if gonorrhea suspected
  6. Virus-holding medium, i.e., M4 medium for the transport of viruses and Chlamydia(Remel, Lenexa, KS, USA)
  7. Chlamydiaculture transport medium, i.e., M4 medium for the transport of viruses and Chlamydia (Remel, Lenexa, KS, USA)
  8. Sabourad's agar if fungal conjunctivitis suspected
  9. Technique
  10. Method for staining the cornea for epithelial defect
  11. Instill a very small amount of fluorescein in the lower conjunctival fornix by lightly touching the tear film with a fluorescein strip. Flooding the eye with fluorescein may obscure a small corneal epithelial defect.
  12. Evaluate the cornea for staining with a Wood lamp or other blue light source.
  13. If a corneal epithelial defect is present, the cornea may be infected and an ophthalmologist should be consulted.
  14. Herpes virus may present in the neonate as a geographic-shaped epithelial effect rather than a dendrite.

FIG. 23.1. Everting the upper eyelid.

  1. Method for everting eyelids
  2. Upper lid (Fig. 23.1)
  3. Grasp lashes and border of lid between thumb and index finger of nondominant hand.
  4. Draw lid downward and away from eyeball.
  5. Indent upper lid, with handle of cotton-tipped applicator held in dominant hand and pull lid back and upward over applicator.
  6. Remove applicator and hold lid in place with nondominant hand by gently pressing border of lid against superior orbital margin.
  7. Lower lid (Fig. 23.2)
  8. Place index finger of nondominant hand on margin of lower lid.
  9. Pull downward.
  10. Method for obtaining cultures

Obtain cultures prior to conjunctival scraping. Take separate cultures from each eye with a separate sterile


swab for each type of medium desired. Culture and label each eye separately, even if only one eye is symptomatic. The uninfected eye can serve as a control for indigenous flora (8).


FIG. 23.2. Using Kimura platinum spatula to take scraping from lower eyelid.

  1. Moisten calcium alginate swabs with trypticase soy broth or other liquid culture medium.
  2. Evert eyelid.
  3. Apply swab to bulbar and palpebral conjunctiva of upper and lower fornices of eye.
  4. Apply swab directly to culture medium plates at the bedside with a single row of C-shaped inoculation streaks. Monitoring the growth of organisms along the shape of the inoculation streaks may help the laboratory in the diagnosis of the cultured pathogen.
  5. Use a separate sterile swab for each culture plate or culture vial.
  6. Label cultures meticulously with eye cultured (right or left) and part of eye cultured (conjunctiva, lid margin, etc.).
  7. Incubate cultures immediately.
  8. Method for obtaining conjunctival scrapings for smear and nonculture Chlamydiatests

 .   Evert eyelid as described above.

  1. Instill topical anesthetic into conjunctival fornix, if desired.
  2. Swab off excess discharge.
  3. Take scraping 2 mm from eye margin. (Normal keratinized epithelium from the lid margin may confound results of smear.)
  4. Pass spatula two to three times in the same direction, avoiding bleeding.
  5. Spread specimen from spatula gently into a monolayer on a clean glass slide and label.
  6. Fix smears as required for proposed smears and nonculture Chlamydiatests.
  7. Repeat with separate sterile spatula on second eye.
  8. Interpretation of Conjunctival Cytology
  9. Cellular reaction
  10. Polymorphonuclear reaction
  11. Bacterial infections
  12. Chlamydial infection
  13. Very severe viral infection
  14. Mononuclear reactions: viral infection
  15. Eosinophilia and basophilia: allergic states
  16. Plasma cells: chlamydial infection
  17. Intraepithelial cell inclusions
  18. Chlamydial infection
  19. Acidophilic inclusions in cytoplasm, capping epithelial cell nuclei
  20. Basophilic “initial bodies” in cytoplasm
  21. Viral infection

Giant, multinucleated epithelial cells may be seen (e.g., herpetic keratoconjunctivitis).

  1. Complications of Scraping
  2. Conjunctival bleeding
  3. Mild conjunctival bleeding, usually self-limiting, frequently occurs.
  4. Instill erythromycin ophthalmic ointment.
  5. Corneal injury
  6. Keep the spatula blade flat against the tarsal conjunctiva at all times to avoid trauma to the cornea.
  7. Corneal injury is confirmed by a staining defect on fluorescein staining.
  8. If corneal injury occurs, instill erythromycin ophthalmic ointment and contact an ophthalmologist.
  9. Transfer of infection from infected to noninfected eye

This complication is avoided by using separate sterile instruments when taking samples from each eye.

  1. Ocular irritation, pain, photophobia, lacrimation, swelling, and hyperemia

These problems are usually mild and self-limited.


  1. De Toledo AR, Chandler JW.Conjunctivitis of the newborn. Infect Dis Clin North Am. 1992;6:807–813.
  2. World Health Organization.Conjunctivitis of the Newborn: Prevention and Treatment at the Primary Health Care Level. Geneva: World Health Organization; 1986:1–23.
  3. Haas J, Larson E, Ross B, et al. Epidemiology and diagnosis of hospital acquired conjunctivitis among neonatal intensive care unit patients. Pediatr Infect Dis J.2005;24(7):586–589.
  4. Faden H, Wynn RJCampagna L, et al. Outbreak of adenovirus type 30 in the neonatal intensive care unit. J Pediatr.2005;146(4):523–527.
  5. Casolari C, Pecorari M, Fabio G, et al. A simultaneous outbreak of Serratia marcescensand Klebsiella pneumoniae in a neonatal intensive care unit. J Hosp Infect. 2005;61(4): 312–320.
  6. Hammerschlag MR.Neonatal conjunctivitis. Pediatr Ann. 1993;22:326–361.
  7. O'Hara MA.Ophthalmia neonatorum. Pediatr Clin North Am. 1993;40:715–725.
  8. Miller JM, ed. A Guide to Specimen Management in Clinical Microbiology.2nd ed. Washington, DC: American Society for Microbiology Press; 1999.
  9. Mullin GS, Rubinfeld RS.The antibacterial activity of topical anesthetics. Cornea. 1997;16:662–665.



  1. Benson WH, Lanier JD.Comparison of techniques for culturing corneal ulcers. Ophthalmology. 1992;99:800–804.
  2. Jacob P, Gopinathan U, Sharma S, et al. Calcium alginate swab versus Bard Parker blade in the diagnosis of microbial keratitis.Cornea.1995;14:360–364.
  3. Rapoza PA, Johnson S, Taylor HR.Platinum spatula vs Dacron swab in the preparation of conjunctival smears [Letter]. Am J Ophthalmol. 1986;102:400–412.
  4. Talley AR, Garcia-Ferrer FLaycock KA, et al. Comparative diagnosis of neonatal chlamydial conjunctivitis by polymerase chain reaction and McCoy cell culture. Am J Ophthalmol.1994;117: 50–57.
  5. Hammerschlag MR, Roblin PM, Gelling M, et al. Use of polymerase chain reaction for the detection of Chlamydia trachomatisin ocular and nasopharyngeal specimens from infants with conjunctivitis. Pediatr Infect Dis J. 1997;16:293–297.
  6. Kowalski RP, Thompson PP, Kinchington PR, et al. Evaluation of the Smart Cycler II system for real-time detection of viruses andChlamydiafrom ocular specimens. Arch Ophthalmol. 2006;124(8): 1135–1139.
  7. Chichili GR, Athmanathan S, Farhatullah S, et al. Multiplex polymerase chain reaction for the detection of herpes simplex virus, varicella-zoster virus and cytomegalovirus in ocular specimens. Curr Eye Res.2003;27(2):85–90.