Emergence of penicillin-resistant Streptococcus pneumoniae in southern Ontario , 1993-94

Departments of Microbiology and Medicine, Sunnybrook Health Science Centre, University of Toronto; Med-Chem Laboratories; and Flemingdon Medical Laboratory, Toronto, Ontario Correspondence and reprints: Dr A Simor, Department of Microbiology, B121, Sunnybrook Health Science Centre, 2075 Bayview Avenue, North York, Ontario M4N 3M5. Telephone (416) 480-4549, Fax (416) 480-6845 Received for publication December 13, 1994. Accepted February 10, 1995 AE SIMOR, A RACH LIS, L LOUIE, J GOOD FEL LOW, M LOUIE. Emer gence of penicillinresistant Strep to coc cus pneu mo niae in south ern On tario, 19931994. Can J In fect Dis 1995;6(3):157160.

P ENI CIL LIN HAS LONG BEEN CON SID ERED TO BE THE ANTIBIOTIc of choice for the treat ment of in fec tions due to Strep tococ cus pneu mo niae.In the past few years, though, there has been a world wide in crease in the preva lence of penicillinresistant S pneu mo niae (1)(2)(3).Al though a strain of pneu mococ cus with re duced sus cep ti bil ity to peni cil lin was first reported in Can ada 20 years ago (4), since then there have been only spo radic re ports of in va sive in fec tions due to re sistant or gan isms in this coun try (5)(6)(7).Three large sur veys of pneu mo coc cal sus cep ti bil ity in Can ada found rates of re sistance to peni cil lin of 2.4%, 1.3% and 1.5% in Al berta (8), Quebec (9) and On tario (10), re spec tively.Two re cent cases of in va sive in fec tion (men in gi tis, bac tere mic pneu mo nia) due to penicillin-resistant strains of S pneu mo niae seen at our hos pital prompted us to con duct a pi lot study to de ter mine the preva lence of re sis tance of S pneu mo niae to peni cil lin and other an ti mi cro bial agents in met ro poli tan To ronto.We describe here one of the cases and the re sults of the preva lence sur vey.

CASE PRES EN TA TION
A 45-year-old male with hu man im mu no de fi ciency vi rus infec tion and cen tral nerv ous sys tem lym phoma was ad mit ted to hos pi tal in No vem ber 1993 with a two-day his tory of fe ver and dysp nea.His medi ca tions on ad mis sion in cluded trimethoprim-sulfamethoxazole, flu cona zole and predni sone.Physi cal ex ami na tion re vealed cel lu li tis over the an te rior aspect of his neck, and a chest x-ray re vealed left lower lobe con soli da tion.Blood and spu tum cul tures grew S pneu moniae re sis tant to peni cil lin (mini mal in hibi tory con cen tra tion [MIC] 4 µg/mL).He was ad mit ted to the in ten sive care unit with res pi ra tory fail ure.He was treated with van co my cin for two weeks and was dis charged from hos pi tal.How ever, he was re ad mit ted one month later with re cur rent fe ver and bac teremia due to penicillin-resistant S pneu mo niae.He was retreated with van co my cin, but died sev eral months later of causes un re lated to pneu mo coc cal in fec tion.

PA TIENTS AND METH ODS
Con secu tive S pneu mo niae iso lates from dif fer ent pa tients were ob tained be tween June and De cem ber 1993, and between March and June 1994.Iso lates were ob tained from Med-Chem Labo ra to ries and Fleming don Medi cal Labo ra tory, two pri vate community-based la borato ries pro vid ing serv ices to fam ily phy si cians and nurs ing homes in met ro poli tan To ronto, and from pati ents as sessed in the emer gency de part ment of a ter ti ary-care teach ing hos pi tal in To ronto.All iso lates were identi fied as S pneu mo niae based on co lo nial mor phol ogy, Gram stain char ac ter is tics, op to chin sus cep ti bil ity and bile solu bility.In vi tro sus cep ti bil ity test ing was done by a broth mi cro dilu tion pro ce dure, fol low ing Na tional Com mit tee for Clini cal Labo ra tory Stan dards guide lines (11).An in ter me di ate level of re sis tance to peni cil lin was de fined as MIC 0.1 to 1.0 µg/mL; high-level re sis tance was de fined as an MIC 2.0 µg/mL or greater (11).
Pneu mo cocci were se ro typed by the Na tional Ref er ence Cen tre for Strep to coc cus (Ed mon ton, Al berta).Ge nomic DNA of re sis tant iso lates was also ex am ined by pulsed field gel elec tro pho re sis (PFGE).Ge nomic DNA from pneu mo coc cal isolates was pre pared in agar plugs as pre vi ously de scribed (12) and di gested with SmaI (Boehringer-Mannheim).Di gested DNA was elec tro pho re sed us ing the contour-clamped ho moge ne ous elec tric field ap pa ra tus (CHEF-DRII, Bio Rad, Califor nia).Elec tro pho re sis was car ried out for 20 h us ing ramped pulse times be gin ning with 0.2 s and end ing with 25 s, at an ap plied volt age of 6 V/cm.The gels were stained with ethidium bro mide and pho to graphed un der ul tra vio let il lu mi na tion.Iso lates were con sid ered to rep re sent dif fer ent strains if the PFGE band pat terns dif fered by at least three bands.
(30%) of which had high-level re sis tance and 14 with in ter medi ate re sis tance (Ta ble 1).Four teen re sis tant strains were eye, ear or spu tum iso lates from pe di at ric out-patients and six strains (two from spu tum and one each from blood, cere brospi nal fluid, bron choal veo lar wash ings and the eye) were from adults seen in the emer gency de part ment.Penicillinsusceptible strains were gen er ally sus cep ti ble to the other anti mi cro bial agents tested.How ever, penicillin-resistant S pneu mo niae were also fre quently re sis tant to cef tazidime (55%), tet ra cy cline (55%), trimethoprim-sulfamethoxazole (50%), eryth ro my cin (40%), ce fu r ox ime (35%) and cef tri axone (25%) (Ta ble 1).Iso lates were uni formly sus cep ti ble to only van co my cin and imipe nem.
The penicillin-resistant pneu mo cocci were se ro typed as fol lows: 19F (six iso lates), 9V (three), 23F (three), and one each of 6A, 6B, 14, 19A; four were non type able.The PFGE results of a rep re sen ta tive sam ple of peni cil lin-r esi stant isolates are shown in Fig ure 1. Dif fer ent se ro types had clearly dis tin guish able PFGE pat terns.How ever, whereas all three sero type 9V iso lates had iden ti cal pat terns and a non type able strain shared an iden ti cal pat tern with a se ro type 23F iso late, the other 23F se ro types had dis tinct pro files and each of the 19F se ro type iso lates had a unique pat tern.

DIS CUS SION
The re cent in crease in the preva lence of peni cil lin-r esistant S pneu mo niae in the met ro poli tan To ronto re gion found in this sur vey is in marked con trast to re sults ob tained in a 1988 study (10) of a simi lar community-based popu la tion when only eight of 551 (1.5%) penicillin-resistant strains were de tected (P<0.0001).Fur ther more, none of the pre vi ously iden ti fied pneu mo coc cal iso lates had dem on strated highlevel peni cil lin re sis tance, whereas 30% of the re sis tant strains in this sur vey had high-level re sis tance.
The re cent in crease in preva lence is also re flected by no tifi ca tions of in va sive pneu mo coc cal dis ease due to penicillin-resistant strains to the On tario Pneu mo coc cal Study Group (13) in 1993-94.
The re sults of se ro typ ing and mo lecu lar typ ing by PFGE sug gest that mul ti ple clones of penicillin-resistant pneu mococci are ap pear ing si mul ta ne ously in the met ro poli tan Toronto re gion.Simi lar re sults have been sug gested by a cross-Canada sur vey of penicillin-resistant iso lates (14).Moreo ver, this pre limi nary ex pe ri ence with these typ ing methods sug gests that for S pneu mo niae, PFGE is more dis crimi natory than is se ro typ ing; this ob ser va tion would have to be con firmed by evalu at ing a larger number of iso lates.
The emer gence of penicillin-resistant S pneu mo niae has ma jor health care im pli ca tions.Mi cro bi ol ogy labo ra to ries must be able to de tect rap idly and ac cu rately peni cil lin re sistance in pneu mo coc cal iso lates.Al though drug-resistant strains of pneu mo cocci now ap pear to be much more prevalent in sev eral parts of Can ada, re gional varia tions may ex ist.

TA BLE 1 An ti mi cro bial sus cep ti bil ity of 274 iso lates of Strep to coc cus pneu mo niae
If the re sults of this sur vey are con firmed else where in Canada, em piri cal an ti mi cro bial treat ment of a va ri ety of in fectious dis eases (in clud ing pneu mo nia, bron chi tis, oti tis me dia, si nusi tis and men in gi tis) will have to be modi fied.Treat ment op tions may be lim ited be cause penicillin-resistant strains are also fre quently re sis tant to other beta-lactam an ti bi ot ics (15,16).Moreo ver, al ter na tive an ti mi cro bial agents for the treat ment of pneu mo coc cal in fec tions may not be as ef fec tive as peni cil lin would be for sus cep ti ble strains (17,18).Be cause in va sive in fec tions due to re sis tant strains are as so ci ated with con sid er able mor bid ity and mor tal ity, greater use of pneu mococ cal vac cine should be pro moted, par ticu larly for those at high risk for se vere pneu mo coc cal in fec tions (19).Con tinuous sur veil lance of S pneu mo niae an ti mi cro bial sus cep ti bil ity is re quired to de ter mine the ex tent of re sis tance to peni cil lin and al ter na tive agents, so that ap pro pri ate ther apy and control meas ures may be in sti tuted.

An ti mi cro bial agent Penicillin-susceptible iso lates (n=254) Penicillin-resistant iso lates (n=20)
*MIC90 Mini mal in hibi tory con cen tra tion of an ti mi cro bial agent re quired to in hibit growth of 90% of the iso lates testedFig ure1) Re stric tion en do nu cle ase analy sis of ge nomic DNA of penicillin-resistant Strep to coc cus pneu mo niae iso lates ob tained by pulsed field gel elec tro pho re sis af ter di ges tion with SmaI.Lanes 1 and 2, se ro type 9V iso lates from two dif fer ent pa tients; lane 3, non typeable iso late; lane 4, se ro type 23F; lanes 5-7, se ro type 19F iso lates from three dif fer ent pa tients; lanes 8-10, se ro types 6A, 6B and 19A, re spec tively