| Philbrick JT et al, 1988, USA | All studies of sufficient quality identified from literature search over years 1944 to 1986 | Literature review | Incidence PE | 6 of 163 patients receiving no anticoagulation (Strength of evidence weak). 9 of 208 patients receiving a range of anticoagulation (0 of 32 in only study with strong evidence, all other studies, weak) | |
| Giachino A, 1988, Canada | 152 patients who died in Ottawa hospitals over a five year period, with PE listed as the cause of death. | Retrospective study | Source of thrombosis in fatal pulmonary emboli. | 82 had no postmortem examination. 23 postmortem examinations confirmed PE as the cause of death, and identified the source of the embolus. 3 of 23 postmortem examinations revealed the calf veins as the source of the thrombi. | No controlling of postmortem procedures—unclear if all legs veins thoroughly examined. |
| | | | | Only 23 of 152 considered to die from PE actually had a PM and had the source of the embolus confirmed. |
| Lohr J et al, 1991, USA | 75 patients with ultrasound diagnosed calf thrombosis. Treatment left to physician’s discretion. | Prospective study with follow up serial ultrasound examination. | Thrombosis propagation | 15% propagated to involve the popliteal or larger veins. A further 17% propagated within the calf veins | Publication bias—all of these patients may have been included in the study by Pelligrini V et al, 1993. |
| | | | | No information regarding the length of follow up, or the effect of varying therapies |
| Pellegrini V et al, 1993, USA | 25 patient with isolated calf DVT and 12 patients with superficial or muscular calf thrombosis, diagnosed by venography on postoperative screening of total hip arthroplasty patients. Only 12 calf DVTs and one superficial/muscular calf thrombosis were anticoagulated | Prospective study following up at 6, 12, 24 and 52 weeks | Incidence of PE | 4 of 13 untreated calf DVT patients were diagnosed with PE. 0 of 1 treated calf DVT patient and none of the superficial/muscular calf thrombosis developed PE | Two of the PEs were diagnosed on the strength of sudden collapse and cardiac arrest—no postmortem examination carried out |
| Nielson HK et al, 1994, Denmark | 15 patients with venographically diagnosed calf DVTs. | Prospective study | VQ scan result at presentation | 5 of 15 had positive VQ scans | No information regarding exact criteria for diagnosing PE from VQ scan alone—probable over-estimation of incidence |
| | | | | VQ scans were performed at 10 and 60 days, however no information regarding the breakdown of subsequent PEs between proximal and isolated calf DVT groups |
| Lohr JM et al, 1995, USA | 192 patients with ultrasound diagnosed below knee DVTs, Treatment left to physicians discretion | Prospective study with serial ultrasound for four weeks | Thrombus propagation | 53 of 139 thrombi propagated | Publishing bias—the cohort appears to include all of the patients included in the previous Lohr study (see study in this table) |
| | | | | Paper does not establish rate of PE |
| O’Shaughnessy AM et al, 1997, Ireland | 50 patients with ultrasound diagnosed DVTs, 43 treated with anticoagulation and 7 without | Prospective study, using repeat ultrasound at one week, one month, six months and one year. | “Outcome” of isolated calf thrombosis. | 3 patients presented initially with a “positive” VQ scan. One fatal PE within the first month. | Venography not used to diagnose initial calf DVT. Apparently, no attempts were made to actively seek the diagnosis of PE throughout the follow up period. No adequate description of the positive VQ scans. 10 patients lost to follow up at six months. No account taken of the effect of treatment |
| Gottlieb RH et al, 1999, USA | 238 patients with ultrasound diagnosed below knee DVTs | Retrospective study | Incidence of diagnosed PEs | 2 of 56 patients not receiving anticoagulant therapy had PE | Patients were not identified using venography. Retrospective study, therefore unable to detect silent PEs or those that did not present to medical services. One PE diagnosed on strength of high probability VQ scan alone. No description of frequency of follow up ultrasound scans. Therapy at the discretion of physician. No information regarding anticoagulant therapy for patient with extension to thigh DVT. 28 patients were not followed up for the full six months as they died |
| | | Incidence of extension into thigh DVT | 1of 227 receiving anticoagulant therapy had documented extension to thigh DVT | |
| Pinede L et al, 2001, France | 105 patients with calf DVTs treated for six weeks with warfarin, 92 patients with calf DVTs treated for 12 weeks with warfarin | Prospective study | Incidence of PE | 1 of 197 (patient from 12 week warfarin group) had documented PE | Diagnosis did not always use venography |
| | | | | No information regarding which symptoms would prompt investigations for PE. Method’s description implies that a VQ scan result of intermediate probability would diagnose PE—no information as to how this PE was diagnosed |
| Schwarz T et al, 2001, Germany | 84 patients with isolated calf muscle thrombosis. 52 received LMWH for 10 days, 32 received no anticoagulation | Prospective cohort with serial ultrasound examinations | Progression to deep veins of calf | Study discontinued as 8 of 32 non-anticoagulated patients progressed to deep veins thrombosis, compared with 0 of 52 anticoagulated patients | Gold standard venography not used |
| | | | | VQ scan results interpreted in isolation |
| | | PE | None | |
| Sharpe RP et al, 2002, USA | 85 trauma patients with below knee DVTs | Prospective cohort | Thrombus propagation | 4 of 85 thrombi propagated proximally | Gold standard investigations not applied for DVT or PE |
| | | PE | 1 of 85 did not propagate but had a PE | |